Fri 1 May 2026
Daily Brief — Curated and contextualised by Best Practice AI
US Economy Grows, Huawei Gains, and Nvidia Faces Price Surge
TL;DR The US economy grew by 2% in Q1, driven by AI investments, rebounding from a previous government shutdown. Huawei's chip sales are set to jump 60% as Chinese tech firms move away from Nvidia. OpenAI's new model release spurs debate over computing power, highlighting its 10-gigawatt capacity milestone. Nvidia's B300 server prices have nearly doubled in China due to strong demand and a crackdown on chip smuggling.
The stories that matter most
Selected and contextualised by the Best Practice AI team
The Impact of LLM Self-Consistency and Reasoning Effort on Automated Scoring Accuracy and Cost
arXiv:2604.26954v1 Announce Type: new Abstract: Strategic model selection and reasoning settings are more effective than ensembling for optimizing automated scoring with large language models (LLMs). We examined self-consistency (intra-model majority voting) and reasoning effort for scoring conversation-based assessment items in high school mathematics, evaluating 900 student conversations against human-scored ground truths using frontier and low-cost models from OpenAI and Google. Temperature sampling significantly improved accuracy over deterministic calls, but increasing ensemble size (j = 1 to 7) produced no significant gains. Higher reasoning effort showed a significant positive linear trend with scoring accuracy, though the benefit varied by model family. An efficiency frontier analysis identified Gemini 3.1 Pro Preview at low reasoning as the most accurate but costly configuration; GPT-5.4 Nano and Mini with no reasoning offered the best cost-performance balance.
OpenAI’s New Model Spurs Debate Over Computing Power
Sam Altman suggested it would be released more widely than a rival offering from Anthropic. Some are suggesting it’s because OpenAI has more computing power.
OpenAI Reaches 10-Gigawatt AI Capacity Milestone Years Ahead of Target - Bloomberg
OpenAI Reaches 10-Gigawatt AI Capacity Milestone Years Ahead of Target - Bloomberg OpenAI: BackForward Gift this article Contact us:Provide news feedback or report an error Confidential tip?Send a tip to our reporters Site feedback:Take our Survey April 30, 2026 at 12:47 AM UTC Save Translate OpenAI has met a key milestone for securing AI capacity in the US several years ahead of schedule, boosting the startup’s ambitious plans for data center expansion. The ChatGPT creator has signed contracts for 10 gigawatts of artificial intelligence computing capacity, it said in a blog post on Wednesday. The company had originally aimed to reach that goal by 2029. Before it's here, it's on the Bloomberg Terminal LEARN MORE News Work & Life Market Data Explore Terms of ServiceDo Not Sell or Share My Personal InformationTrademarksPrivacy Policy CareersAdvertise Ad Choices Help©202
EU Chips Act Overhaul Aims to Boost Investment - Bloomberg
EU Chips Act Overhaul Aims to Boost Investment - Bloomberg Chip Wars: US AI Export Control BackForward Semiconductor wafer fabrication in Germany. Photographer: Krisztian Bocsi/Bloomberg Gift this article Contact us:Provide news feedback or report an error Confidential tip?Send a tip to our reporters Site feedback:Take our Survey By Gian Volpicelli and Alberto Nardelli April 30, 2026 at 12:34 PM UTC Corrected April 30, 2026 at 2:53 PM UTC Save Translate The European Union’s revamped plans to stimulate the semiconductor industry — a crucial part of the supply chain for everything from artificial intelligence to cars — would allow the bloc’s executive arm to invest directly in manufacturing and would prioritize development of new technologies, people familiar with the draft said. The proposal for the Chips Act II, expected in late May, is an attempt to improve on its 2022
Council Post: Why The Economics Of AI Are Nothing Like SaaS
When the cost to serve is variable, the pricing has to be variable too.
So, About That AI Bubble
Thanks to the rise of Claude Code and other AI agents, revenues are finally catching up to the hype.
Elon Musk Acknowledges Partial Distillation of OpenAI Models in xAI Grok Training – Firstpost
Elon Musk’s testimony in the OpenAI lawsuit highlighted debates over AI model distillation, industry competition, and Open Ai's pivot from its non profit mission
For What AI Could Do to Democracies, Look to the Petrostates
Societies will become richer, but history suggests that wealth may not be equally distributed.
The Secret Weapon Against AI Dominance
The future of creative labor will turn on whether AI-generated work can be copyrighted.
Unpacking Vibe Coding: Help-Seeking Processes in Student-AI Interactions While Programming
arXiv:2604.27134v1 Announce Type: new Abstract: Generative AI is reshaping higher education programming through vibe coding, where students collaborate with AI via natural language rather than writing code line-by-line. We conceptualize this practice as help-seeking, analyzing 19,418 interaction turns from 110 undergraduate students. Using inductive coding and Heterogeneous Transition Network Analysis, we examined interaction sequences to compare top- and low-performing students. Results reveal that top performers engaged in instrumental help-seeking -- inquiry and exploration -- eliciting tutor-like AI responses. In contrast, low performers relied on executive help-seeking, frequently delegating tasks and prompting the AI to assume an executor role focused on ready-made solutions. These findings indicate that currently generative AI mirrors student intent (whether productive or passive) rather than optimizing for learning. To evolve from tools to teammates, AI systems must move beyond passive compliance. We argue for pedagogically aligned design that detect unproductive delegation and adaptively steer educational interactions toward inquiry, ensuring student-AI partnerships augment rather than replace cognitive effort.
Divergence Between Frontier Model APIs and Native Integrated Applications
A growing performance gap exists between general-purpose model APIs and native applications developed by frontier labs. This suggests that vertical integration of models into specific harnesses provides superior agentic capabilities compared to raw API access.
4 in 10 enterprises see 40%+ productivity gains from AI in customer support: Report - Storyboard18
Kapture CX survey shows 40% enterprises reporting over 40% productivity gains from AI in customer support, with growing automation, uneven adoption, and cost benefits tied to deeper workflow integration.
Generative AI adoption speed unprecedented, O’Reilly survey says | InfoWorld
Generative AI adoption speed unprecedented, O’Reilly survey says | InfoWorld # Generative AI adoption speed unprecedented, O’Reilly survey says news Nov 22, 20232 mins ## Survey of enterprise users of generative AI finds rapid adoption but also hurdles, with difficulty finding business use cases, legal uncertainties, and high infrastructure costs top concerns. Credit: Thinkstock Generative AI, the wave of artificial intelligence led by OpenAI’s GPT large language models and ChatGPT, is experiencing rapid, never-before-seen levels of adoption, according to a report from technology publisher and training provider O’Reilly. But issues remain with adoption, including lack of perceived business cases and worrisome legal questions. The company’s report, 2023 Generative AI in the Enterprise, published November 21, said two-thirds of survey respondents already were using generative AI. “W
Hidden IT problems are quietly creating risk, shadow IT, and lost productivity
Presented by TeamViewer Enterprise technology failures are largely invisible. Research from TeamViewer, based on a global survey of 4,200 managers and employees, finds that the majority of digital dysfunction never reaches the IT help desk. Employees work around slow applications, failed logins, and intermittent glitches rather than reporting them, leaving organizations without an accurate picture of how their technology is performing. The cumulative cost is significant: employees lose an average of 1.3 workdays per month to digital friction, with impacts ranging from delayed projects and lost revenue to increased employee turnover. The research, which surveyed managers and employees across nine countries, confirms what many have long suspected: the productivity loss from digital friction is significant, and most of it never surfaces in an IT support queue, says Andrew Hewitt, VP of strategic technology at TeamViewer. “Enterprise outages are visible because they trigger clear, system-level failures,” Hewitt says. “But much of the real disruption happens earlier, in the form of digital friction: slow apps, login issues, or intermittent glitches that don’t cross alert thresholds. These smaller issues often go unreported or are normalized by employees, even though they quietly drain productivity.” What is digital friction and why does it go unreported? The most common sources of friction — connectivity failures, software crashes, hardware problems, and authentication issues — aren’t edge-case scenarios, but everyday experiences employees have learned to absorb without escalating. Connectivity problems were the most widespread, with nearly half identifying them as the top productivity killer among common technology issues. That tendency to absorb rather than report is central to the problem. Many workers don’t trust their IT team to resolve issues quickly or effectively, so when a login fails or an application stalls mid-task, the path of least resistance is to restart the device, switch tools, or use a personal phone. “Employees are under more pressure than ever to prove output,” Hewitt says. “When reporting feels unlikely to result in a quick resolution, it creates a false sense of stability at the system level while the employee experience quietly deteriorates.” How much productivity does digital friction cost organizations? The business consequences extend beyond inconvenience. Many organizations report delays in critical operations, revenue loss, and lost customers as a result of IT dysfunction. Most respondents lose time each month, and few expect improvement, citing increasing complexity of workplace technology as a primary concern. The human cost runs parallel. Workers link digital friction to frustration, decreased motivation, and burnout, and many believe it contributes to turnover, with onboarding replacements stretching to eight weeks or more. "Employees are happiest when they feel productive and accomplished at the end of the day," Hewitt says. "When people can't make progress in their day-to-day work, frustration builds and burnout follows. Great technology might not be a main attractor of talent, but bad technology can certainly play a role in driving it away." Why employees use personal devices and unauthorized tools instead of reporting IT problems When workplace technology consistently fails to meet employee needs, workers find alternatives, with a substantial share of respondents admitting to using personal devices or unauthorized applications as workarounds. That's the entry point for shadow IT, or the use of unapproved hardware, software, or cloud services outside IT's visibility and control. While employees turn to these tools simply to stay productive, they introduce security vulnerabilities, data leakage risks, and compliance gaps that IT teams may not discover until a breach occurs. “Quite simply, it demonstrates that the IT environment is not meeting the employees’ needs,” Hewitt said. “While this helps maintain short-term productivity, it introduces significant risks and pushes work outside of IT’s visibility and control.” TeamViewer ONE addresses this by combining remote connectivity with real-time endpoint monitoring, giving IT teams the ability to detect and resolve device and application issues before employees reach for an alternative. When the underlying environment is stable and support is fast, the impulse to work around it diminishes. How fragmented IT infrastructure creates blind spots across devices, apps, and networks Addressing digital friction at scale requires more than faster help desk response times. Traditional metrics such as mean time to resolution and ticket volume capture only a fraction of actual issues. A more complete picture requires measuring lost time, interrupted workflows, and employee sentiment across devices, applications, and network environments. “Leaders need to move beyond measuring performance through IT tickets alone,” Hewitt said. “Performance should be viewed through the lens of employee experience and real-time digital workplace data.” Fragmented infrastructure makes this difficult. When devices, applications, and networks operate in separate silos, IT teams struggle to trace root causes or identify systemic issues before they spread, often responding to symptoms rather than underlying problems. TeamViewer ONE is designed to close that gap, integrating digital employee experience analytics, remote support, and device management into a single platform. Instead of piecing together signals from disconnected tools, IT teams get a consolidated view of endpoint health, application performance, and network conditions across the entire organization. How organizations can shift from reactive IT support to proactive system monitoring Achieving proactive IT is not a single-step transformation. Hewitt describes it as a progression: starting with endpoint management and security, building toward real-time visibility into the digital employee experience, and ultimately using automation and AI to resolve issues before they reach employees. TeamViewer AI is built to support each stage of that progression, using continuous monitoring to surface anomalies and correlate signals across the digital environment, identifying patterns of poor experience before they escalate. When issues are detected, it suggests remediations, generates scripts to fix problems autonomously, and handles routine tasks such as common troubleshooting without requiring IT intervention, shifting the workload from reactive firefighting toward proactive oversight. And while AI's effectiveness depends on the completeness of the data it works with, consolidating onto a platform like TeamViewer ONE removes that limitation by giving AI a complete, real-time data foundation to work from. How system performance lays the foundation for productivity, retention, and competitive advantage TeamViewer ONE isn't a wholesale replacement of existing IT infrastructure, but a unifying layer that connects insight with action, which enables organizations to ramp up productivity, improve retention, and ultimately realize a significant competitive advantage. It begins with visibility into what is actually causing friction across their environment. From there, leaders can use that data to prioritize fixes, and then scale remediation through automation as confidence and capability grow. "Reducing digital friction isn't about overhauling everything at once," Hewitt said. "Leaders should start small, gain visibility into what's actually causing friction, fix the biggest pain points, then scale those improvements through automation and AI. Even incremental progress can make an impact on employee engagement and productivity." Dig deeper: Fix it before they feel it from TeamViewer. Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. For more information, contact sales@venturebeat.com.
DVLA's 14-week driving license fiasco – the tech, people and chatbot trying to clear it
Medical license applicants still waiting months while agency insists it's 'putting things right' The Driver and Vehicle Licensing Agency (DVLA) has introduced new techto support driving license applications that require medical checks, after processing times exceeded 14 weeks in February.…
How to figure out if AI is making you more productive - Fast Company
AI makes you feel like you're more productive, but how can you actually figure out if it's worth using?
Italy closes antitrust probes into AI firms after commitments ... - Reuters
Italy closes antitrust probes into AI firms after commitments on 'hallucination' risks | Reuters Exclusive news, data and analytics for financial market professionalsLearn more aboutRefinitiv AI (Artificial Intelligence) letters are placed on computer motherboard in this illustration taken, June 23, 2023. REUTERS/Dado Ruvic/Illustration Purchase Licensing Rights, opens new tab MILAN, April 30 (Reuters) - Italy's antitrust authority said on Thursday it had closed investigations into three AI companies over allegedly unfair commercial practices involving generative artificial intelligence, after accepting binding commitments from them. The regulator, known as the AGCM, also polices consumer rights. It said it had targeted China's DeepSeek, France's Mistral AI SAS and Turkey's Scaleup Yazilim Hizmetleri Anonim Şirketi over risks of so-called AI hallucinations - the generation of
Rep. Dan Goldman on AI companies: 'Regulate them all'
# Rep. Dan Goldman on AI companies: 'Regulate them all' Published: 2026-04-30T18:26:41+00:00 ## Summary Rep. Dan Goldman, who is facing a primary challenge from former New York City comptroller Brad Lander, answered questions about AI and data center regulation, his support for U.S. military aid for Israel, and his understanding of the situation. The Post's Anna Liss-Roy is asking candidates for their candid candidacies regarding these issues. ## Story Rep. Dan Goldman on AI companies: 'Regulate them all' (Anna Liss-Roy/The Washington Post) Up next in Politics Politics # Rep. Dan Goldman on AI companies: 'Regulate them all' April 30, 2026 | 6:16 PM GMT Rep. Dan Goldman (D-New York), who faces a primary challenge from former New York City comptroller Brad Lander, answered rapid-fire questions on AI and data center regulation, his support of U.S. military aid for Israel and what sets
Economics & Markets
Council Post: Why The Economics Of AI Are Nothing Like SaaS
When the cost to serve is variable, the pricing has to be variable too.
Dan Ives Sees Apple ‘Golden Age’ as It Monetizes Consumer AI
“The consumer AI revolution is gonna go through Cupertino,” says Dan Ives, global head of tech research and senior equity analyst at Wedbush Securities, as he discusses expectations for Apple’s AI strategy after the company delivered a surprisingly strong revenue forecast for the third quarter. (Source: Bloomberg)
So, About That AI Bubble
Thanks to the rise of Claude Code and other AI agents, revenues are finally catching up to the hype.
AI companies are just companies
As we leap into a new technological age, the old rules of capitalism still apply
The Artificial Investor #77: OpenAI strategic pivot to Enterprise agents, Meta forces once again a utility/data privacy tradeoff and the business case for a 60-billion-dollar AI acquisition
The Artificial Investor #77: OpenAI strategic pivot to Enterprise agents, Meta forces once again a utility/data privacy tradeoff and the business case for a 60-billion-dollar AI acquisition # The Artificial Investor SubscribeSign in # The Artificial Investor #77: OpenAI strategic pivot to Enterprise agents, Meta forces once again a utility/data privacy tradeoff and the business case for a 60-billion-dollar AI acquisition Apr 30, 2026 Share My name is Aris Xenofontos and I am an investor at Seaya Ventures. This is the weekly version of the Artificial Investor that goes deep into a topic from the top AI developments of the last days. Three stories this week crystallize the inflection points defining AI’s next phase: OpenAI’s enterprise agent pivot threatens to commoditise the very category Anthropic pioneered, Meta’s keystroke logging exposes the privacy tradeoff that could trigger
Companies accelerating digital investments, but struggling to demonstrate their business impact - study - Business Review
Companies accelerating digital investments, but struggling to demonstrate their business impact - study - Business Review # Companies accelerating digital investments, but struggling to demonstrate their business impact – study Anca Alexe 30/04/2026 | 12:10 Companies are accelerating investments in digital transformation, but struggle to clearly demonstrate the value generated by these initiatives, according to the study “Digital Value: How AI and Digital Transformation Finally Pay Off” by international management consulting company Horváth. The findings show a strong increase in digitalisation budgets, but also highlight gaps in strategy and performance measurement. Key findings of the study: - 67% of companies are increasing their digitalisation budgets, with an average rise of around 30% - Around 30% of investments are directed towards artificial intelligence (AI) initiatives -
Apple’s R&D spending hits new record as AI investment ramps up - 9to5Mac
Apple’s spending on research and development hit an all-time high in the company’s latest quarterly results, reflecting an industry-wide continued push into AI development.
The AI Boom Is Driving GDP Growth - What’s News - WSJ Podcasts
# The AI Boom Is Driving GDP Growth - What’s News - WSJ Podcasts Published: 2026-04-30T20:52:00+00:00 Type: Opinion ## Summary The article discusses the impact of the AI on the U.S. on its way to the world's consciousness. The article also discusses the growing population of the country, with an estimated 6.9.10. . The authorises the publication of the article. ## Story P.M. Edition for April 30. for the first three months of the year, rising from the previous quarter but not as fast as economists were expecting. , chief economist at EY-Parthenon, joins to discuss the business investments fueling that growth. Plus, U.S. national debt now . Hear from Journal investing columnist on how that could affect government activity. And the latest tech giant reports: and revenue top Wall Street expectations. Alex Ossola hosts. Sign up for the WSJ's free .
Tax refunds and AI boom have offset some US economic pain from ...
Tax refunds and AI boom have offset some US economic pain from Iran war and high gas prices, so far - The Washington Post Democracy Dies in Darkness By Paul Wiseman and Christopher Rugaber | AP WASHINGTON — Americans are paying for the war in Iran with every visit to the gas station, but some of the damage to the U.S. economy is being offset — for now anyway — by big tax refunds and an investment boom driven by artificial intelligence.
Musk testifies he did not read 'fine print' about OpenAI ... - Reuters
Musk testifies he did not read 'fine print' about OpenAI becoming for-profit company | Reuters Exclusive news, data and analytics for financial market professionalsLearn more aboutRefinitiv Item 1 of 6 OpenAI attorney William Savitt cross-examines Elon Musk as his deposition is played on a screen, during Musk's lawsuit trial over OpenAI's for-profit conversion before U.S. District Judge Yvonne Gonzalez Rogers, with Sam Altman, CEO of OpenAI, sitting in the foreground, at a federal courthouse in Oakland, California, U.S., April 30, 2026, in a courtroom sketch. REUTERS/Vicki Behringer [1/6]OpenAI attorney William Savitt cross-examines Elon Musk as his deposition is played on a screen, during Musk's lawsuit trial over OpenAI's for-profit conversion before U.S. District Judge Yvonne... Purchase Licensing Rights, opens new tab Read more - Summary - Companies - Musk alleges OpenAI priorit
Why does EU want Google to open up Android to AI rivals? - The Hindu
Why does EU want Google to open up Android to AI rivals? - The Hindu You are logged in Loading... LOGOUT You don’t have any Active Subscription. Subscribed with another email? Logout and Login with that one. Your active subscription(s) Account subscription benefits alongside Premium Stories, Editorials, Opinions and more. Unlock these with Subscription Products you've access to Additional Subscription Benefits Account Settings Need help with your subscription? # Why does EU want Google to open up Android to AI rivals? ## The European Commission has shared draft measures that it wants Google to implement, in order to let AI rivals better access Android’s key capabilities Published - April 30, 2026 10:16 am IST READ LATER Under the current system, the European Commission observed that Google was favouring its own AI offerings (namely Gemini) on Android devices [File] | Phot
Netomi raises $110 million as Accenture and Adobe bet on AI for customer service
Netomi, the San Francisco-based startup building AI systems for enterprise customer service, said Thursday that it has raised $110 million in new funding in a round led by Accenture Ventures, with participation from Adobe Ventures, WndrCo, Silver Lake Waterman, NAVER Ventures, Metis Strategy and Fin Capital. Jeffrey Katzenberg, managing partner of WndrCo and co-founder of DreamWorks, has joined the company's board. The round builds on early backing from a roster of AI luminaries that includes OpenAI co-founder Greg Brockman, Google DeepMind co-founder Demis Hassabis and Microsoft AI CEO Mustafa Suleyman. On its face, the financing is another large AI round in a market still awash in capital. But the deal is more revealing than that. It suggests that a new line is being drawn inside enterprise AI — not between companies that have a chatbot and companies that do not, but between companies that can show AI works in the messy, brittle, heavily governed environments where large businesses actually operate, and those that still mostly shine in demos. The market around Netomi makes the stakes clear. Sierra, the AI agent startup led by former Salesforce co-CEO Bret Taylor, raised $350 million at a $10 billion valuation in September 2025 and has since made three acquisitions in 2026 alone. Decagon tripled its valuation to $4.5 billion in January 2026 with a $250 million Series D. Salesforce, ServiceNow and Intercom are all racing to embed AI agents into their existing platforms; Intercom's Fin AI agent reportedly crossed $100 million in annual recurring revenue at $0.99 per resolution. Gartner predicts that 40 percent of enterprise applications will include task-specific AI agents by the end of 2026, up from less than 5 percent in 2025. Against that backdrop, Netomi's $110 million round is not the largest in the category, but it may be the most strategically constructed. The combination of Accenture's enterprise consulting network, Adobe's dominance in digital experience management and Netomi's track record in production deployments represents a coordinated play to embed AI not as a chatbot layer on top of websites, but as the fundamental intelligence governing how entire digital experiences behave. The company did not disclose its valuation, and in an interview tied to the announcement, Netomi executives declined to provide revenue or profitability figures. Instead, Chief Executive Puneet Mehta pointed to customer economics, saying a typical large deployment can generate at least tens of millions of dollars in impact, with some customers on a path to hundreds of millions. For technical decision-makers, though, the more important part of Thursday's news may be the partnerships attached to the money. Why Accenture and Adobe turned a venture deal into a global distribution play The structure of the deal reads like a map of how enterprise AI gets bought in 2026. Alongside the investment, Accenture has entered a global alliance with Netomi to bring the platform to its Fortune 100 client base worldwide. The alliance will involve hundreds of Accenture team members receiving training on Netomi's platform — a meaningful commitment from the world's largest consulting firm and a distribution channel that few AI startups can match. Adobe Ventures' participation comes with plans to integrate Netomi into Adobe's Brand Concierge agentic ecosystem, giving Netomi a path into the software layer many large brands already use to manage websites, content and digital journeys. Metis Strategy brings access to CIO advisory channels. Ndidi Oteh, CEO of Accenture Song, said in the press release that the partnership is designed to help clients "reinvent how they serve their customers — seamlessly, responsibly and at scale." The result is not just more cash. It is a distribution network wrapped around a thesis. Justin Wexler, a partner at WndrCo who led the firm's Series B investment in Netomi in 2021, said most companies in the customer experience space are simply swapping a human for an AI. "That's the extent of what they're building," Wexler said. "What we're doing at Netomi, particularly with the Adobe partnership, is leapfrogging that altogether — merging the two layers. You don't have a 'How can I help you?' chatbot. This is anticipating the issue and eliminating the ticket altogether." The distinction matters because it describes a fundamentally different kind of product. Most customer service AI still sits downstream. A customer encounters a problem, opens a chat window, explains the issue and waits for a response. Even when AI speeds up that exchange, the friction has already happened. Netomi wants to move upstream, into the experience before the ticket exists. Mehta described the idea in blunt economic terms. "Why are there so many customer service tickets? Why is $500 billion spent on human labor answering customer service phone calls, emails and chats?" he asked. "What we realized is that the world's largest companies wait for a problem to happen and then jump on it to solve it — but by that time, they've already created a lot of frustration, and it's very expensive to do that." The answer, in Mehta's view, is not to make downstream customer service faster with AI. It is to prevent the service ticket from being created in the first place. That logic sits behind almost every strategic decision the company has made — including the Adobe partnership. "Most important websites run on Adobe Experience Manager," Mehta said. "So we're saying, what if we bring that kind of context and awareness upstream — capturing that a customer might be affected before it even turns into a customer service ticket." The Wall Street trading floor origins behind Netomi's AI architecture To understand what Netomi is building, you have to understand where its founder came from. Mehta, who spent his early career constructing automated trading engines on Wall Street, told VentureBeat that the founding thesis was deceptively simple. "When we started Netomi, the core thesis was that AI is going to become the new customer interface," he said. "The Transformers [paper] did not exist, so we had literally stitched together a set of different models to create the same end result." That background in low-latency finance is not incidental. It is the intellectual architecture that undergirds everything Netomi builds. When asked what connects trading systems to customer experience platforms, Mehta drew a direct line. "If you think about the low-latency trading world, that was the first technology application to use situational awareness and a variety of different signals at scale," he said. "There was not one signal that it was making decisions on. You needed market data feeds. You needed situational awareness. You needed news. You needed awareness of your own book of business. You needed your own risk assessment." That multi-signal architecture, Mehta argued, translates directly to what enterprise customer experience demands. Rather than waiting passively for a customer to describe a problem — the way traditional chatbots and even most current AI agents operate — Netomi's system attempts to reconstruct the full situation before it acts. The request itself is only part of the story. "What the customer tells you is very important, but the situation the customer is in is sometimes even more important," Mehta said. "What if we borrowed that design pattern we built for low-latency trading? Because we can probably know why the customer is calling us. And if we can know that, we could maybe even reach out to them before they reach out to us and solve the problem." He summarized the philosophical distinction this way: "What large language models by themselves did was they essentially democratized just raw intelligence. We are democratizing context, and that changes everything." That is a sharp line, and also a revealing one. Netomi is effectively betting that the defensible layer in enterprise AI will not be the foundation model alone. It will be the orchestration layer that turns general model capability into governed, auditable, domain-specific action. That governed approach extends to how the platform handles risk. Netomi uses what it calls an AI authority matrix — a real-time system that defines what the AI can do autonomously and when it must escalate to a human. "It's a little bit like autonomous driving," Mehta said. The AI knows when it's approaching a boundary and pulls a human in. For regulated industries, specific endpoints can be locked to deterministic, rules-based flows while the agentic layer handles broader orchestration — and all of it is version-controlled and traceable, with metadata saved for seven years. Inside the AI system that rearranges websites and retail stores in real time The most technically ambitious element of Netomi's vision — and the one that most sharply distinguishes it from competitors — is what the company calls AI-embedded customer experience orchestration. Rather than placing a chatbot in the corner of a website, Netomi's system can rearrange the website itself based on what the AI infers about each individual customer's situation. Wexler demonstrated a live example during the interview. "As we see most deployments, companies that want to deploy AI on their websites, they throw a chatbot on the corner," he said. "If you embed agentic capabilities into the digital layer itself — and again, Adobe Experience Manager is the leading digital layer of enterprise — then you could do really unique things." Wexler described what this looks like in practice. In a typical deployment, he said, the AI doesn't just answer questions — it reshapes the page. Based on a customer's browsing behavior, purchase history and inferred intent, the system can reorganize a product page in real time: surfacing warnings one customer needs but another doesn't, prompting a sample order at the moment of hesitation, or flagging a compatibility issue before checkout. Two customers looking at the same product might see fundamentally different pages — not because a marketing team built two versions, but because the AI is composing the experience on the fly. "The AI is playing the role of arranging the elements of the website to cater to me and my needs," Wexler said. "It's anticipating my needs." The implication is a shift from static web pages to something closer to generative websites — pages that reconstruct themselves around each visitor the way a good salesperson adjusts a pitch mid-conversation. It is a fundamentally different model from bolting a chat widget onto a page that otherwise looks the same for everyone. That vision already extends beyond screens. Mehta revealed that Coach, the handbag company owned by Tapestry, deployed Netomi's platform in a physical flagship store during the holiday season to help customers navigate the retail space and is now rolling it out chainwide. The numbers Netomi is putting behind its production claims are equally ambitious. At DraftKings, the company said its platform can handle traffic surging to more than 40,000 concurrent customer requests per second during major sporting events, while delivering sub-three-second response times and 98 percent intent classification accuracy. At Paramount, the company said it deployed across chat and voice in two weeks and then scaled through a weekend that included a major UFC event and the AFC Championship. Those are company-reported numbers, and they are hard to benchmark against competitors because the industry lacks standard public reporting. But they illustrate the kind of problem Netomi wants buyers to think about. At that scale, an AI support product stops looking like a smarter FAQ bot and starts looking like a distributed systems challenge. You are not just asking whether a model can answer a question. You are asking whether an entire system can make decisions quickly, safely and consistently while traffic spikes and business rules collide. The $110 million question: can invisible AI beat the chatbot industrial complex? Whether Netomi can deliver on the full scope of its ambition — transforming from an AI customer service platform into an ambient intelligence layer that reshapes digital and physical experiences in real time — remains an open question. The company faces competitors with far larger war chests, deeper platform footprints and, in Sierra's case, a founder-level relationship with OpenAI. But Netomi's bet is fundamentally different from what much of the field is building. While Sierra and Decagon race to replace human agents with AI concierges, measuring success in conversations handled, Netomi is wagering that the highest form of customer service is the interaction that never needs to happen at all. "There are new startups trying to convince enterprises that if every customer gets a 'concierge,' if there's 'an agent for every moment,' then loyalty follows," Mehta said. "But most relationships with brands are functional. Customers don't want a conversational relationship with their airline or their bank. They want things to work — seamlessly, invisibly, without friction." In his closing comments during the interview, Mehta warned that many companies still underestimate the operational risk of deploying immature AI into sensitive customer environments. "What large companies adopting AI don't fully realize yet is what kind of risk are they taking by adopting those platforms that are not really field tested for this kind of scale and situations," he said. That may be the most important line in the whole announcement. Because beneath the funding round, beneath the partner logos and beneath the talk of agents and orchestration, the real question in enterprise AI remains old-fashioned: which systems can be trusted when the environment gets ugly? "We have built this technology more like how automated trading got built, or how autonomous driving got built, compared to coming at this from just a customer service lens," Mehta said. It is a fitting frame for a company whose founder left Wall Street to fix customer service. On the trading floor, the best systems were never the ones that made the most trades. They were the ones that knew, with precision, when not to act — and the ones nobody noticed until something went wrong and they held. Netomi's new investors are betting $110 million that the same principle applies when the person on the other end of the system is not a trader, but a customer who just wants their floor not to leak.
Thiel’s Founders Fund Raises $6 Billion in Its Largest-Ever Haul
Peter Thiel’s Founders Fund has raised $6 billion for a new fund to invest in later-stage companies, according to people familiar with the matter, marking the firm’s largest haul ever.
AI customer service startup Netomi raises $110 million | Reuters
California-based Netomi draws on such technology, using AI models from Open AI , Anthropic and Alphabet’s (GOOGL.O), opens new tab Google, Mehta said.
“If Figma and Lovable had a child that became an architect”: Synaps raises €3.06 million to rival AutoCAD
Five months after launching its first beta version, Vienna-based AI canvas for architects, Synaps, has announced its €3. 06 million ($3. 6 million) pre-Seed investment round from US-based Plug and Play and Fil Rouge, among others.
An early-stage VC shares why he's not investing in AI coding startups — but doubling down on these founders
Jussi Salovaara, managing partner at startup accelerator Antler Asia, said he doesn't see much more room for new entrants in the AI-assisted coding space.
Algorithms Are Deciding Which Startups Get Seen By Venture Capitalists or Not
Thomas Hellmann, researcher and ... says, “AI systems have effectively become a gatekeeping infrastructure between sourced deals and partner attention, fundamentally changing how investment pipelines operate.” The research is based on qualitative analysis of venture capital firms, including ...
Mindflair Expands into Gaming Through SVV Investment in AI Testing Start-up ManaMind
Mindflair plc (LSE:MFAI), the AIM-listed artificial intelligence investment company, has gained indirect exposure to the gaming industry following a new investment by Sure Valley Ventures’ second fund in ManaMind, a London-based start-up specialising in autonomous game testing.
Labor, Society & Culture
It’s time to tax AI slop | Mike Pepi | The Guardian
We are stuck in a deluge of meaningless content that threatens human creativity. Here’s a simple way to mitigate its harms
The Proliferation of AI-Generated Content and Its Impact on Professional Communication
Frequent AI users are increasingly able to identify synthetic prose, creating a new cultural divide in digital communication. This trend suggests a potential devaluation of standard AI-generated text in professional and social environments.
The Impact of AI-Generated Text on the Internet
arXiv:2604.26965v1 Announce Type: new Abstract: The proliferation of AI-generated and AI-assisted text on the internet is feared to contribute to a degradation in semantic and stylistic diversity, factual accuracy, and other negative developments (sometimes subsumed under the Dead Internet Theory). What has hindered answering these questions is that it has not been understood just how much of the internet is actually AI-generated or AI-edited. To this end, we construct a representative sample of websites published on the internet between 2022 and 2025 using the Internet Archive, and apply a state-of-the-art AI text detector on them. We find that by mid-2025, roughly 35% of newly published websites were classified as AI-generated or AI-assisted, up from zero before ChatGPT's launch in late 2022. We also find statistically significant evidence for some of the identified hypotheses; for example, that increases in AI-generated text on the internet correlate negatively with semantic diversity and positively with the prevalence of positive sentiment. We do not, however, find statistically significant evidence supporting the hypothesis that an increased rate of AI-generated text on the internet decreases factual accuracy or stylistic diversity. Notably, this diverges from public perception, which we measure in a user study, where the majority of US adults turned out to believe in all four of the above-mentioned hypotheses. Individuals who do not use AI or use it infrequently tend to believe in these negative impacts more than those who use it frequently; similarly, individuals who hold negative views of AI tend to believe in these hypotheses more than those with favorable views of the technology.
Can AI be a moral victim? The role of moral patiency and ownership perceptions in ethical judgments of using AI-generated content
arXiv:2604.26956v1 Announce Type: new Abstract: The growing use of generative AI raises ethical concerns about authorship and plagiarism. This study examines how people judge the reuse of AI-generated content, focusing on moral patiency and ownership perceptions. In an experiment, participants evaluated two substantively similar manuscripts in which the original source was described as authored by a human, an AI system, or an AI agent with a human-like name. Results showed that copying AI-generated work was judged less unethical, less plagiaristic, and less guilt-inducing than copying human-authored work. Mediation analyses revealed that this leniency stemmed from lower perceptions of AI's capacity to suffer harm (moral patiency) and greater ownership attributed to the human writer reusing AI-generated content. Anthropomorphic cues shaped moral evaluations indirectly by reducing perceived ownership. These findings shed light on how people morally disengage when using AI-generated work and highlight differences in how ethical judgments are applied to human versus AI-created content.
Reporters at McClatchy Withhold Bylines in A.I. Dispute
Journalists at newspapers like The Miami Herald and The Sacramento Bee are refusing to let the chain use their names on summarized articles generated by a new A.I. tool.
‘Awkward and humiliating’: UK job hunters share frustration with AI interviews
People describe unnatural process as survey finds nearly half of job seekers have been interviewed by AI Nearly half (47%) of UK job seekers have had an AI interview, research from the hiring platform Greenhouse has found. In its survey of 2,950 active job seekers, including 1,132 UK-based workers, with additional respondents from the US, Germany, Australia and Ireland, it found that 30% of UK candidates had walked away from a hiring process because it included an AI interview. Continue reading...
Inside India newsletter: AI is exposing cracks in India’s growth story as it hits high-paying IT jobs
Inside India newsletter: AI is exposing cracks in India’s growth story as it hits high-paying IT jobs Key Points - 10 million to 15 million people working in the information technology sector have anchored India's aspirational middle class—buying homes, taking flights, driving consumption. - For the last five years, gross hiring of IT firms averaged around 230,000, but in the financial year ending in March 2026, they added around 170,000: analyst. - Ten-plus years of 'Make in India' has not yet triggered a manufacturing renaissance, an expert said. This report is from this week's "Inside India" newsletter, which brings you timely, insightful news and market commentary on the emerging powerhouse — Subscribe today Hello, this is Priyanka Salve, writing to you from Singapore. Welcome to the latest edition of " Inside India" — your one-stop destination for stories and developments from
Tesla's former HR chief: the AI layoff panic Is built on a false premise—here's what most workers need to know | Fortune
The hyperscaler math driving cuts at Meta and Microsoft does not apply to the companies where most Americans actually work.
The Secret Weapon Against AI Dominance
The future of creative labor will turn on whether AI-generated work can be copyrighted.
From grading papers to decoding jargon, here are some ways ...
From grading papers to decoding jargon, here are some ways people are putting AI to work - The Washington Post Democracy Dies in Darkness By Cathy Bussewitz | AP NEW YORK — Artificial intelligence is permeating workplaces, changing the nature of jobs of every stripe. Teachers are using it to create lesson plans and grade papers. Marketing professionals are harnessing it to work a room and learn about the needs of potential clients. Product managers are asking AI to serve as an interpreter when technical conversations went over their heads in meetings.
No indication of AI displacing jobs widely, as firms using AI see productivity gains: MOM report - CNA
No indication of AI displacing jobs widely, as firms using AI see productivity gains: MOM report - CNA #### Recent Searches #### Trending Topics Advertisement Advertisement # No indication of AI displacing jobs widely, as firms using AI see productivity gains: MOM report Early evidence suggests AI is complementing rather than displacing labour in Singapore, though AI adoption at work still remains low. Office workers at Raffles Place in Singapore. (File photo: Marcus Mark Ramos) New: You can now listen to articles. This audio is generated by an AI tool. ###### Davina Tham 30 Apr 2026 06:00PM (Updated: 30 Apr 2026 06:11PM) Set CNA as your preferred source on Google Add CNA as a trusted source to help Google better understand and surface our content in search results. Read a summary of this article on FAST. Get bite-sized news via a newcards interface. Give it a try. Click
Productivity Commission wrong on AI jobs impact - MacroBusiness
Late last month, Productivity Commission (PC) chair Danielle Wood downplayed the impact of artificial intelligence (AI) on Australia’s job market, claiming only around 4% of jobs are at risk of elimination: “They [Jobs and Skills Australia] found a reasonably small share of jobs, about ...
4 in 10 Students Say AI Will Influence Their Career Choice
4 in 10 Students Say AI Will Influence Their Career Choice April 30, 2026 # 4 in 10 Students Say AI Will Influence Their Career Choice Students report feeling “uncertain,” “concerned,” “nervous” and “depressed” about how AI will impact their future careers, a new survey finds. Also, high cost of living deters them from pursuing a college degree. By Asked how they feel about AI’s impact on their future career, half of student respondents answered, “Uncertain.” Photo illustration by Justin Morrison/Inside Higher Ed | PhonlamaiPhoto/iStock/Getty Images | ruizluquepaz/E+/Getty Images Nearly half—42 percent—of college-eligible students say that artificial intelligence will influence which career they pursue, and 10 percent report that they have already changed their planned major due to concerns about AI, according to a report released Tuesday. “AI is upending the value equation in hi
Ethical AI Governance Framework for Enterprises: Key Principles and Steps
Rapid AI adoption has outpaced oversight, creating governance gaps that expose enterprises to risk, liability, and trust erosion. An ethical AI governance framework establishes principles, roles, and lifecycle controls ensuring fairness, transparency, accountability, privacy, and safety.
Governance for safe and responsible AI in healthcare organisations: a scoping review of frameworks | npj Digital Medicine
This scoping review synthesises current evidence on artificial intelligence (AI) governance in healthcare organisations, outlining key components of AI governance frameworks. Following PRISMA-ScR guidelines, we searched MEDLINE, Embase, and Scopus (April 2024, updated March 2025) for AI governance ...
How bots could help revive democracy
New AI tools encouraging deliberation could promote consensus and reduce polarisation in politics
In praise of tech troublemakers
As CEOs refuse constraints, workers feel a responsibility to prevent AI’s most dangerous uses
Unpacking Vibe Coding: Help-Seeking Processes in Student-AI Interactions While Programming
arXiv:2604.27134v1 Announce Type: new Abstract: Generative AI is reshaping higher education programming through vibe coding, where students collaborate with AI via natural language rather than writing code line-by-line. We conceptualize this practice as help-seeking, analyzing 19,418 interaction turns from 110 undergraduate students. Using inductive coding and Heterogeneous Transition Network Analysis, we examined interaction sequences to compare top- and low-performing students. Results reveal that top performers engaged in instrumental help-seeking -- inquiry and exploration -- eliciting tutor-like AI responses. In contrast, low performers relied on executive help-seeking, frequently delegating tasks and prompting the AI to assume an executor role focused on ready-made solutions. These findings indicate that currently generative AI mirrors student intent (whether productive or passive) rather than optimizing for learning. To evolve from tools to teammates, AI systems must move beyond passive compliance. We argue for pedagogically aligned design that detect unproductive delegation and adaptively steer educational interactions toward inquiry, ensuring student-AI partnerships augment rather than replace cognitive effort.
Tech For Good - Beyond free initiatives: Rethinking the UK’s AI talent strategy
Faye Ellis, Principal Training Architect at Pluralsight, explores why the UK’s AI talent strategy must evolve beyond free courses to deliver measurable business impact and close the skills gap.
Technology & Infrastructure
Think it, Run it: Autonomous ML pipeline generation via self-healing multi-agent AI
arXiv:2604.27096v1 Announce Type: new Abstract: The purpose of our paper is to develop a unified multi-agent architecture that automates end-to-end machine learning (ML) pipeline generation from datasets and natural-language (NL) goals, improving efficiency, robustness and explainability. A five-agent system is proposed to handle profiling, intent parsing, microservice recommendation, Directed Acyclic Graph (DAG) construction and execution. It integrates code-grounded Retrieval-Augmented Generation (RAG) for microservice understanding, an explainable hybrid recommender combining multiple criteria, a self-healing mechanism using Large Language Model (LLM)-based error interpretation and adaptive learning from execution history. The approach is evaluated on 150 ML tasks across diverse scenarios. The system achieves an 84.7% end-to-end pipeline success rate, outperforming baseline methods. It demonstrates improved robustness through self-healing and reduces workflow development time compared to manual construction. The study introduces a novel integration of code-grounded RAG, explainable recommendation, self-healing execution and adaptive learning within a single architecture, showing that tightly coupled intelligent components can outperform isolated solutions.
Govern your bots carefully or chaos could ensue
Stop the sprawl! With the average Global Fortune 500 enterprise expected to run more than 150,000 AI agents by 2028, up from fewer than 15 today, there’s plenty of room for chaos. Analyst firm Gartner says that, without proper governance, those agents will multiply and run amok.…
How to Build a Self-Improving AI Agent Pipeline Using Open Source: Simulate, Evaluate, and Optimize
How to Build a Self-Improving AI Agent Pipeline Using Open Source: Simulate, Evaluate, and Optimize # Future AGI SubscribeSign in # How to Build a Self-Improving AI Agent Pipeline Using Open Source: Simulate, Evaluate, and Optimize ### Build an AI agent pipeline that catches its own failures and rewrites its own prompts. Step-by-step guide using simulate-sdk, ai-evaluation, and agent-opt. Apr 30, 2026 Share Your eval suite passes at 100%. Your agent still fails in production. That contradiction is the starting point of this entire guide. If that sounds familiar, you don’t have a model problem. You have a pipeline problem, and a closed loop of Simulate, Evaluate, and Optimize fixes it. Key Takeaways A 100% passing eval suite usually means your test coverage is broken, not that your agent is ready for production. Tool-call sequence bugs only surface under sentiment-loaded inputs,
How agentic AI transforms platform operations
Ahead of the Innovate Roadshow in Singapore, Dynatrace’s APJ VP & CTO Rafi Katanasho highlights that public sector IT is moving from reactive ‘screen watching’ to defining strategic guardrails and improving service quality.
What is agentic AI and should HR be using it?
Software companies are promoting agentic AI alongside their longstanding HR systems, but what does it do and does HR need it?
Stripe Product Roadmap | What We’re Building Next | Patrick Collison | 27 comments
Stripe Product Roadmap | What We’re Building Next | Patrick Collison | 27 comments Agree & Join LinkedIn By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy. # Patrick Collison’s Post 9h We just announced a large raft of improvements at @Stripe Sessions. My meta reflections: • It feels that the entire economy is replatforming right now. • Many charts at Stripe are inflecting in quite dramatic ways. What GitHub recently reported for commits we are seeing in economic activity (such as new company formations). • It is increasingly clear that agents will be responsible for most transactions in the not overly distant future. • Stripe was always developer-centric, but AI is making developer-centricity strategic in a new way: agents are even hungrier for good DX than developers themselves are. • Things that we’re launching are
Step-level Optimization for Efficient Computer-use Agents
arXiv:2604.27151v1 Announce Type: new Abstract: Computer-use agents provide a promising path toward general software automation because they can interact directly with arbitrary graphical user interfaces instead of relying on brittle, application-specific integrations. Despite recent advances in benchmark performance, strong computer-use agents remain expensive and slow in practice, since most systems invoke large multimodal models at nearly every interaction step. We argue that this uniform allocation of compute is fundamentally inefficient for long-horizon GUI tasks. Such trajectories are highly heterogeneous: many steps are routine and can be handled reliably by smaller, cheaper policies, while errors tend to concentrate at a relatively small number of high-risk moments. Across computer-use benchmarks, these failures repeatedly take two forms: progress stalls, where the agent loops, repeats ineffective actions, or fails to make meaningful progress, and silent semantic drift, where the agent continues taking locally plausible actions after already deviating from the user's true goal. To address this inefficiency, we propose an event-driven, step-level cascade for computer-use agents that runs a small policy by default and escalates to a stronger model only when lightweight learned monitors detect elevated risk. Our framework combines two complementary signals: a Stuck Monitor that detects degraded progress from recent reasoning-action history and triggers recovery, and a Milestone Monitor that identifies semantically meaningful checkpoints where sparse verification is most informative for catching drift. This design turns always-on frontier-model inference into adaptive, on-demand compute allocation over the course of an evolving interaction. The framework is modular and deployment-oriented: it can be layered on top of existing computer-use agents without changing the underlying agent architecture or retraining the large model.
Alibaba's Metis Agent Cuts Redundant AI Tool Calls
Alibaba's Metis agent cuts redundant AI tool calls from 98% to 2% — and gets more accurate doing it.
EU Chips Act Overhaul Aims to Boost Investment - Bloomberg
EU Chips Act Overhaul Aims to Boost Investment - Bloomberg Chip Wars: US AI Export Control BackForward Semiconductor wafer fabrication in Germany. Photographer: Krisztian Bocsi/Bloomberg Gift this article Contact us:Provide news feedback or report an error Confidential tip?Send a tip to our reporters Site feedback:Take our Survey By Gian Volpicelli and Alberto Nardelli April 30, 2026 at 12:34 PM UTC Corrected April 30, 2026 at 2:53 PM UTC Save Translate The European Union’s revamped plans to stimulate the semiconductor industry — a crucial part of the supply chain for everything from artificial intelligence to cars — would allow the bloc’s executive arm to invest directly in manufacturing and would prioritize development of new technologies, people familiar with the draft said. The proposal for the Chips Act II, expected in late May, is an attempt to improve on its 2022
Satya Nadella Said "Exploit" On an Earnings Call. He Meant Every Word. -- AI Brief April 30
Satya Nadella Said "Exploit" On an Earnings Call. He Meant Every Word. -- AI Brief April 30 # Artificially Intimidating SubscribeSign in # Satya Nadella Said "Exploit" On an Earnings Call. He Meant Every Word. -- AI Brief April 30 ### Today's Context Window: OpenAI lands on Amazon Bedrock, Nadella will "exploit" his AI deal, the Senate introduces the CHATBOT Act, and Europe's AI deadline hasn't moved. Apr 30, 2026 Share Good morning, humans. Last night Satya Nadella told a room full of Wall Street analysts that Microsoft “fully plans to exploit” the new OpenAI deal — the same day OpenAI showed up on Amazon’s cloud, 48 hours after leaving Microsoft’s exclusive arrangement. Meanwhile, Big Tech collectively disclosed $130 billion in quarterly AI capex and pledged $700 billion for the full year, which is the GDP of the Netherlands going into data centers. We also have a bipartisan Sen
Microsoft Breaks Cloud Maturity Pattern With Surprise Rebound - FourWeekMBA
Microsoft’s Azure cloud platform achieved 40% year-over-year growth in Q2 2024, defying industry expectations and breaking the traditional pattern where large cloud providers experience inevitable growth deceleration as their revenue bases mature. The software giant’s cloud infrastructure ...
United States AI in Cybersecurity Market to hit US$ 46.77 Billion by 2033, Driven by Increasing Cyber Threat Complexity and Adoption of AI-Powered Security Solutions
In the AI in cybersecurity market, NVIDIA plays a major role by providing high-performance computing platforms and AI frameworks that enable real-time threat detection, anomaly detection, and deep learning-based security analytics. Its technologies are widely used in security operations centers (SOCs), cloud security platforms, and enterprise ...
OpenAI’s New Model Spurs Debate Over Computing Power
Sam Altman suggested it would be released more widely than a rival offering from Anthropic. Some are suggesting it’s because OpenAI has more computing power.
OpenAI Reaches 10-Gigawatt AI Capacity Milestone Years Ahead of Target - Bloomberg
OpenAI Reaches 10-Gigawatt AI Capacity Milestone Years Ahead of Target - Bloomberg OpenAI: BackForward Gift this article Contact us:Provide news feedback or report an error Confidential tip?Send a tip to our reporters Site feedback:Take our Survey April 30, 2026 at 12:47 AM UTC Save Translate OpenAI has met a key milestone for securing AI capacity in the US several years ahead of schedule, boosting the startup’s ambitious plans for data center expansion. The ChatGPT creator has signed contracts for 10 gigawatts of artificial intelligence computing capacity, it said in a blog post on Wednesday. The company had originally aimed to reach that goal by 2029. Before it's here, it's on the Bloomberg Terminal LEARN MORE News Work & Life Market Data Explore Terms of ServiceDo Not Sell or Share My Personal InformationTrademarksPrivacy Policy CareersAdvertise Ad Choices Help©202
Building the compute infrastructure for the Intelligence Age
# Building the compute infrastructure for the Intelligence Age | OpenAI Published: 2026-04-30T02:23:21+00:00 ## Summary OpenAI is leading a long-term effort to build the compute foundation required to deliver the benefits of AGI broadly and reliably to consumers, businesses, developers, and governments. The company has already surpassed its goal of securing 10GW of AI infrastructure in the United States by 2029, with over 3GW added in the last 90 days alone. The move is part of OpenAI's strategy to expand its compute footprint and bring new capacity online faster to meet increasing demand for AI. These projects require a combination of power, land, permitting, transmission, workforce, and partner readiness. The latest and smartest model yet, GPT‐5.5, was trained at our flagship Stargate site in Abilene, Texas. ## Story Building the compute infrastructure for the Intelligence Age | O
Nebius Agrees to Buy Startup That Makes AI Run Faster, Cheaper
The cloud provider’s deal for Eigen AI will help the inference business
One Tool Call to Rule Them All
New open source Python tool Runpod Flash eliminates containers for faster AI dev.
AI’s Bottleneck Is Steel, Silicon and Ships
Boris Kriuk is right: AI does not fail when a model misbehaves; it falters when the supply chains beneath it do. The real constraint on AI is no longer clever code, but the physical and institutional systems that make large‑scale computation possible.
Why water will shape the next phase of UK AI data center growth
The race to scale AI infrastructure is accelerating. In the UK, billions are flowing into new data center capacity as operators respond to surging demand for high-performance compute. Much of the discussion has centered on power grid capacity, generation, and how quickly new supply can be brought online.
Mapping Out America’s AI Data Center Boom: 4,000 Sites and 442 GW in Active Queue | TheEnergyMag
A U.S. data center map shows nearly 4,000 sites today, but 442 GW of active interconnection requests reveal the true scale of AI-driven power demand — echoing a buildout first seen in Bitcoin mining.
Meta's 1Q26 earnings redraw the AI hardware map
The debut of the Muse Spark model ... a hardware buildout of historic proportions — one that will ripple across semiconductor fabs, HBM suppliers, server ODMs, and wearable component makers across the region for years to come. At US$56.3 billion in quarterly revenue, Meta is no longer simply a software-first social platform. It is fast becoming one of the most aggressively vertically integrated AI infrastructure buyers in the world — and Asia's supply chain sits squarely ...
Divergence Between Frontier Model APIs and Native Integrated Applications
A growing performance gap exists between general-purpose model APIs and native applications developed by frontier labs. This suggests that vertical integration of models into specific harnesses provides superior agentic capabilities compared to raw API access.
This startup’s new mechanistic interpretability tool lets you debug LLMs
The San Francisco–based startup Goodfire just released a new tool, called Silico, that lets researchers and engineers peer inside an AI model and adjust its parameters—the settings that determine a model’s behavior—during training. This could give model makers more fine-grained control over how this technology is built than was once thought possible. Goodfire claims Silico…
How cyber security is changing in the age of AI
The advantage will go to the organisations that can pivot to understanding that the economics of cyber crime have completely changed
CareGuardAI: Context-Aware Multi-Agent Guardrails for Clinical Safety & Hallucination Mitigation in Patient-Facing LLMs
arXiv:2604.26959v1 Announce Type: new Abstract: Integrating large language models (LLMs) into patient-facing healthcare systems offers significant potential to improve access to medical information. However, ensuring clinical safety and factual reliability remains a critical challenge. In practice, AI-generated responses may be conditionally correct yet medically inappropriate, as models often fail to interpret patient context and tend to produce agreeable responses rather than challenge unsafe assumptions. Unlike clinicians, who infer risk from incomplete information, LLMs frequently lack contextual awareness. Moreover, real-world patient interactions are open-ended and underspecified, unlike structured benchmark settings. We present CareGuardAI, a risk-aware safety framework for patient-facing medical question answering that addresses two key failure modes: clinical safety risk and hallucination risk. The framework introduces Clinical Safety Risk Assessment (SRA), inspired by ISO 14971, and Hallucination Risk Assessment (HRA) to evaluate medical risk and factual reliability. At inference time, CareGuardAI employs a multi-stage pipeline consisting of a controller agent, safety-constrained generation, and dual risk evaluation, followed by iterative refinement when necessary. Responses are released only when both SRA and HRA are less than or equal to 2, ensuring clinically acceptable outputs with bounded latency. We evaluate CareGuardAI on PatientSafeBench, MedSafetyBench, and MedHallu, covering both safety and hallucination detection. Across these benchmarks, the framework consistently outperforms strong baseline models, including GPT-4o-mini, demonstrating the importance of context-aware, risk-based, inference-time safety mechanisms for reliable deployment in healthcare.
Enforcing trust and transparency: Open-sourcing the Azure Integrated HSM | Microsoft Azure Blog
Enforcing trust and transparency: Open-sourcing the Azure Integrated HSM | Microsoft Azure Blog Unpack all things agentic on The Shift podcast: Listen and subscribe. ## Tech Community Connect with a community to find answers, ask questions, build skills, and accelerate your learning. Visit the Azure Infrastructure Blog tech community As cloud workloads become more agentic and AI systems handle increasingly sensitive data, trust must be engineered directly into infrastructure. Azure Integrated HSM brings hardware‑enforced key protection into Azure, extending cryptographic trust from silicon to services through verifiable and transparent design. As cloud workloads become more agentic and AI systems increasingly handle mission‑critical data, trust must be engineered into the infrastructure at every layer. At Microsoft, security is designed into the foundation of our cloud infrastructu
The never-ending supply chain attacks worm into SAP npm packages, other dev tools
Mini Shai-Hulud caught spreading credential-stealing malware The wave of supply chain attacks aimed at security and developer tools has washed up more victims, namely SAP and Intercom npm packages, plus the lightning PyPI package.…
Okta just ran an event around operational risk for federal agencies: synthetic voice and the collapse of voice as an identity signal. Identity is no longer a control point, it's frontline vs gen AI… | Resemble AI
Okta just ran an event around operational risk for federal agencies: synthetic voice and the collapse of voice as an identity signal. Identity is no longer a control point, it's frontline vs gen AI… | Resemble AI Agree & Join LinkedIn By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy. # Resemble AI’s Post 8,791 followers 4h Okta just ran an event around operational risk for federal agencies: synthetic voice and the collapse of voice as an identity signal. Identity is no longer a control point, it's frontline vs gen AI fraud attempts. A 10 second clip can clone a voice well enough to defeat call-center verification, social-engineer a help desk, or impersonate a senior official. Will Krispin had a conversation with Sabrina Lea, Austin A., and Irina Denisenko covering why and how: → deepfake detection and verifying authe
Adoption, Deployment & Impact
Generative AI adoption speed unprecedented, O’Reilly survey says | InfoWorld
Generative AI adoption speed unprecedented, O’Reilly survey says | InfoWorld # Generative AI adoption speed unprecedented, O’Reilly survey says news Nov 22, 20232 mins ## Survey of enterprise users of generative AI finds rapid adoption but also hurdles, with difficulty finding business use cases, legal uncertainties, and high infrastructure costs top concerns. Credit: Thinkstock Generative AI, the wave of artificial intelligence led by OpenAI’s GPT large language models and ChatGPT, is experiencing rapid, never-before-seen levels of adoption, according to a report from technology publisher and training provider O’Reilly. But issues remain with adoption, including lack of perceived business cases and worrisome legal questions. The company’s report, 2023 Generative AI in the Enterprise, published November 21, said two-thirds of survey respondents already were using generative AI. “W
Hitting the ‘GenAI wall’: Where generative AI stops working, and what it means for your talent strategy
Companies that assume GenAI eliminates the need for expertise will hit the AI wall—and wonder why their workforce transformation has stalled.
Slalom targets AI transformation that earns its spend - SiliconANGLE
Slalom targets AI transformation that earns its spend - SiliconANGLE SHARE Coverage from SiliconANGLE's livestreaming video studiohelp_outline UPDATED 10:34 EDT / APRIL 30 2026 AI ### No more waiting: Enterprise AI transformation has become the top CEO mandate SHARE Enterprise AI transformation is clearing the proof-of-concept stage for many organizations, with execution at scale becoming the new challenge that IT departments alone can’t handle. Governance, talent and organizational structure are required to meet C-suite mandates, according to Daniel Prager(pictured, right), global partner development lead for Google Cloud at Slalom Consulting LLC, a global business and technology consulting firm. Working closely with customers navigating the gap between AI ambition and measurable business impact, Slalom sees these same friction points as the forces most likely to slow even the b
From Prototype to Production: Building AI Systems That Actually Ship | by Aditi S | Apr, 2026 | Medium
From Prototype to Production: Building AI Systems That Actually Ship | by Aditi S | Apr, 2026 | Medium Sign up Get app Sign up # From Prototype to Production: Building AI Systems That Actually Ship 15 min read 11 hours ago -- Share An honest guide to multi-agent architectures, connective infrastructure, and the unglamorous work that separates demos from production. Most AI demos show potential as a POC but rarely gets translated to production. The technology genuinely works in the controlled environment of a Jupyter notebook. But somewhere between “look what it can do” and “this runs in production,” something breaks. Usually several things. According to MIT’s MLQ State of AI in Business 2025 report, only 5% of custom enterprise AI tools ever make it into production. MLQ’s State of AI in Business 2025 report found that 95% of custom enterprise AI initiatives fail to reach produc
AI & Compliance Survey 2026: Adoption is high. Governance and controls lag. - Compliance Week
AI & Compliance Survey 2026: Adoption is high. Governance and controls lag. - Compliance Week In this 2026 survey report from Compliance Week and konaAI, 193 compliance, ethics, risk, and audit leaders reveal how their organizations are deploying AI — and where the gaps are growing. More than 83 percent report using AI tools, yet only about 25 percent have implemented a strong governance framework. Generative AI leads the stack, but data quality issues, lack of expertise, and unmanaged employee use are creating real friction. Meanwhile, executive leadership is driving adoption from the top down, faster than compliance teams can keep up. You must log in or register to access this content. Link Copy link
One year on: Progress on our European digital commitments - Microsoft On the Issues
One year on: Progress on our European digital commitments - Microsoft On the Issues Skip to main content Europe is moving fast to capture the benefits of artificial intelligence, recognizing its potential to raise productivity, strengthen competitiveness, and help modernize public services. At the same time, organizations across Europe are focused on digital sovereignty and resilience: retaining control over their data and critical operations in a period of geopolitical volatility. These priorities go together. That is why one year ago, we announced a set of European digital commitments to respond to these expectations. They focused on five areas: 1. Help build a broad AI and cloud ecosystem across Europe 2. Uphold Europe’s digital resilience even when there is geopolitical volatility 3. Continue to protect the privacy of European data 4. Help protect and defend Europe’s cybersecurit
Council Post: The Great AI Automation Debate: Rising Tides Or Crashing Waves?
AI capabilities are expanding broadly rather than concentrating narrowly. You have time for systematic experimentation, but that time isn't unlimited.
OpenAI’s Big Reset + A.I. in the Doctor’s Office + Talkie, a pre-1930s LLM
Will the rising tide of A.I. adoption lift all boats?
Frontier AI Models Outperform Human Physicians in Clinical Benchmarks and Emergency Scenarios
New research indicates that advanced LLMs can surpass human performance in specific medical diagnostic tasks. These findings underscore an urgent requirement for prospective clinical trials to validate AI efficacy in healthcare settings.
Is AI increasing access to justice?
A steep increase in ‘vibe litigation’ appears to be expanding the market for legal activity
How is AI Voice Assistant Adoption Really Going in 2026? G2's AI Market Report Has the Answer
G2’s 2026 AI voice assistant report shows rapid adoption, proven ROI, and high user satisfaction - despite ongoing ROI measurement challenges.
Carney’s AI push risks harm as Ottawa automates public services
Carney’s AI push risks harm as Ottawa automates public services # The Carney government’s embrace of AI will put lives at risk Deploying these technologies as cost-cutting measures will not only worsen service for Canadians, it will cause physical and mental-health issues. - April 30, 2026 - by Natasha Tusikov Blayne Haggart Image caption:Decision-making algorithms are only as good as their algorithms and the data provided. iStock Share The Mark Carney government has made “ deploying AI at scale” a cornerstone of its attempt to make government more productive and slash costs by cutting 28,000jobs by 2029. The goal is to achieve savings of $60 billion over several years. There are many reasons to be skeptical of the government’s AI strategy. Savings projections resulting from digitalization should be taken with a grain of salt. For example, the Phoenix pay system, designed to auto
Microsoft is delving deeper into professional services workflows with the release of a new AI-powered "Legal Agent" embedded in Microsoft Word, demonstrating how generative AI is transitioning from...
# Microsoft is delving deeper into professional services workflows with the release of a new AI-powered "Legal Agent" embedded in Microsoft Word, demonstrating how generative AI is transitioning from... | Fintech Association Of Kenya Published: 2026-05-01T05:33:39+00:00 ## Summary Microsoft has introduced a new AI-powered "Legal Agent" embedded in Microsoft Word, designed to support legal professionals directly in their work. The agent can help review the contract and provide key insights about its terms and allocation of risk. It can also assist in preparing a fully redefined version of the document. The tool is now available for purchase. ## Story Microsoft is delving deeper into professional services workflows with the release of a new AI-powered "Legal Agent" embedded in Microsoft Word, demonstrating how generative AI is transitioning from... | Fintech Association Of Kenya Agree
DVLA's 14-week driving license fiasco – the tech, people and chatbot trying to clear it
Medical license applicants still waiting months while agency insists it's 'putting things right' The Driver and Vehicle Licensing Agency (DVLA) has introduced new techto support driving license applications that require medical checks, after processing times exceeded 14 weeks in February.…
4 in 10 enterprises see 40%+ productivity gains from AI in customer support: Report - Storyboard18
Kapture CX survey shows 40% enterprises reporting over 40% productivity gains from AI in customer support, with growing automation, uneven adoption, and cost benefits tied to deeper workflow integration.
AI-led process modernization leads to measurable ROI - SiliconANGLE
AI-led process modernization transforms manual invoice approvals into governed, scalable workflows, delivering measurable ROI and operational efficiency.
New AI Tools for Developers 2026: 5 Worth Testing | Medium
New AI Tools for Developers 2026: 5 Worth Testing | Medium Sign up Get app Sign up Press enter or click to view image in full size # 5 New AI Tools for Developers Worth Testing This Month 8 min read 17 hours ago -- Share If you search for new AI tools for developers in 2026, you mostly get the same useless list posts. Fifty tools. Zero point of view. Half of them are wrappers. The other half look impressive for ten minutes and then never make it into your real workflow. I care less about which tool is trending and more about whether it survives contact with an actual project. This month, I kept coming back to five things that feel real enough to test properly. Not because they are perfect, but because they solve a concrete bottleneck in how I ship software. The data backs the urgency. JetBrains’ April 2026 research found that around 90% of developers now use at least one AI t
Solving the Mystery of How Businesses Are Using AI | American Enterprise Institute - AEI
The latest evidence on how companies are using artificial intelligence tells a familiar story: plenty of promise for tomorrow—and modest gains today that are real—but impacts are still hard to see in the big-picture aggregate for companies and the broader US economy.
Hidden IT problems are quietly creating risk, shadow IT, and lost productivity
Presented by TeamViewer Enterprise technology failures are largely invisible. Research from TeamViewer, based on a global survey of 4,200 managers and employees, finds that the majority of digital dysfunction never reaches the IT help desk. Employees work around slow applications, failed logins, and intermittent glitches rather than reporting them, leaving organizations without an accurate picture of how their technology is performing. The cumulative cost is significant: employees lose an average of 1.3 workdays per month to digital friction, with impacts ranging from delayed projects and lost revenue to increased employee turnover. The research, which surveyed managers and employees across nine countries, confirms what many have long suspected: the productivity loss from digital friction is significant, and most of it never surfaces in an IT support queue, says Andrew Hewitt, VP of strategic technology at TeamViewer. “Enterprise outages are visible because they trigger clear, system-level failures,” Hewitt says. “But much of the real disruption happens earlier, in the form of digital friction: slow apps, login issues, or intermittent glitches that don’t cross alert thresholds. These smaller issues often go unreported or are normalized by employees, even though they quietly drain productivity.” What is digital friction and why does it go unreported? The most common sources of friction — connectivity failures, software crashes, hardware problems, and authentication issues — aren’t edge-case scenarios, but everyday experiences employees have learned to absorb without escalating. Connectivity problems were the most widespread, with nearly half identifying them as the top productivity killer among common technology issues. That tendency to absorb rather than report is central to the problem. Many workers don’t trust their IT team to resolve issues quickly or effectively, so when a login fails or an application stalls mid-task, the path of least resistance is to restart the device, switch tools, or use a personal phone. “Employees are under more pressure than ever to prove output,” Hewitt says. “When reporting feels unlikely to result in a quick resolution, it creates a false sense of stability at the system level while the employee experience quietly deteriorates.” How much productivity does digital friction cost organizations? The business consequences extend beyond inconvenience. Many organizations report delays in critical operations, revenue loss, and lost customers as a result of IT dysfunction. Most respondents lose time each month, and few expect improvement, citing increasing complexity of workplace technology as a primary concern. The human cost runs parallel. Workers link digital friction to frustration, decreased motivation, and burnout, and many believe it contributes to turnover, with onboarding replacements stretching to eight weeks or more. "Employees are happiest when they feel productive and accomplished at the end of the day," Hewitt says. "When people can't make progress in their day-to-day work, frustration builds and burnout follows. Great technology might not be a main attractor of talent, but bad technology can certainly play a role in driving it away." Why employees use personal devices and unauthorized tools instead of reporting IT problems When workplace technology consistently fails to meet employee needs, workers find alternatives, with a substantial share of respondents admitting to using personal devices or unauthorized applications as workarounds. That's the entry point for shadow IT, or the use of unapproved hardware, software, or cloud services outside IT's visibility and control. While employees turn to these tools simply to stay productive, they introduce security vulnerabilities, data leakage risks, and compliance gaps that IT teams may not discover until a breach occurs. “Quite simply, it demonstrates that the IT environment is not meeting the employees’ needs,” Hewitt said. “While this helps maintain short-term productivity, it introduces significant risks and pushes work outside of IT’s visibility and control.” TeamViewer ONE addresses this by combining remote connectivity with real-time endpoint monitoring, giving IT teams the ability to detect and resolve device and application issues before employees reach for an alternative. When the underlying environment is stable and support is fast, the impulse to work around it diminishes. How fragmented IT infrastructure creates blind spots across devices, apps, and networks Addressing digital friction at scale requires more than faster help desk response times. Traditional metrics such as mean time to resolution and ticket volume capture only a fraction of actual issues. A more complete picture requires measuring lost time, interrupted workflows, and employee sentiment across devices, applications, and network environments. “Leaders need to move beyond measuring performance through IT tickets alone,” Hewitt said. “Performance should be viewed through the lens of employee experience and real-time digital workplace data.” Fragmented infrastructure makes this difficult. When devices, applications, and networks operate in separate silos, IT teams struggle to trace root causes or identify systemic issues before they spread, often responding to symptoms rather than underlying problems. TeamViewer ONE is designed to close that gap, integrating digital employee experience analytics, remote support, and device management into a single platform. Instead of piecing together signals from disconnected tools, IT teams get a consolidated view of endpoint health, application performance, and network conditions across the entire organization. How organizations can shift from reactive IT support to proactive system monitoring Achieving proactive IT is not a single-step transformation. Hewitt describes it as a progression: starting with endpoint management and security, building toward real-time visibility into the digital employee experience, and ultimately using automation and AI to resolve issues before they reach employees. TeamViewer AI is built to support each stage of that progression, using continuous monitoring to surface anomalies and correlate signals across the digital environment, identifying patterns of poor experience before they escalate. When issues are detected, it suggests remediations, generates scripts to fix problems autonomously, and handles routine tasks such as common troubleshooting without requiring IT intervention, shifting the workload from reactive firefighting toward proactive oversight. And while AI's effectiveness depends on the completeness of the data it works with, consolidating onto a platform like TeamViewer ONE removes that limitation by giving AI a complete, real-time data foundation to work from. How system performance lays the foundation for productivity, retention, and competitive advantage TeamViewer ONE isn't a wholesale replacement of existing IT infrastructure, but a unifying layer that connects insight with action, which enables organizations to ramp up productivity, improve retention, and ultimately realize a significant competitive advantage. It begins with visibility into what is actually causing friction across their environment. From there, leaders can use that data to prioritize fixes, and then scale remediation through automation as confidence and capability grow. "Reducing digital friction isn't about overhauling everything at once," Hewitt said. "Leaders should start small, gain visibility into what's actually causing friction, fix the biggest pain points, then scale those improvements through automation and AI. Even incremental progress can make an impact on employee engagement and productivity." Dig deeper: Fix it before they feel it from TeamViewer. Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. For more information, contact sales@venturebeat.com.
Forget productivity: Here are 5 strategic shifts that drive real AI value | ZDNET
While the CIO I mentioned earlier ... clear, measurable benefits from an AI implementation, Lovelock said executives who don't see the value of extra downtime could be missing a trick. Also: 6 ways to stop cleaning up after AI - and keep your productivity gains · "When people are eight hours head-down, you don't have a corporate ...
Grant Thornton: Insurers See AI Gains but Face Governance Gap
Insurers are seeing gains from artificial intelligence, but many are also exposing themselves to unnecessary risks, according to a new Grant Thornton
Meta: The Superintelligence Wager - LongYield
Advantage+ uses Meta’s first-party engagement data (across Facebook, Instagram, WhatsApp, and Threads) plus AI optimization to allocate advertiser budgets dynamically across campaigns, audiences, creative variants, and placements. Advertisers using Advantage+ have reported 18-32% improvements in return-on-ad-spend (ROAS) and cost-per-acquisition (CPA) versus traditional manual campaigns.
Geopolitics, Policy & Governance
The AP Interview: Ukraine bets on battlefield AI as the race for ...
The AP Interview: Ukraine bets on battlefield AI as the race for weapons autonomy intensifies - The Washington Post Democracy Dies in Darkness By Derek Gatopoulos and Hanna Arhirova | AP KYIV, Ukraine — Rapid military adoption of artificial intelligence is becoming essential to Ukraine’s survival, even as full integration across the battlefield may still be several years away, according to a senior AI official.
India’s AI ambition risks outrunning its readiness | Hindustan Times
Robust data governance frameworks, reforming public procurement to integrate AI responsibly, and digital capability investment within the State are necessities
Italy closes antitrust probes into AI firms after commitments ... - Reuters
Italy closes antitrust probes into AI firms after commitments on 'hallucination' risks | Reuters Exclusive news, data and analytics for financial market professionalsLearn more aboutRefinitiv AI (Artificial Intelligence) letters are placed on computer motherboard in this illustration taken, June 23, 2023. REUTERS/Dado Ruvic/Illustration Purchase Licensing Rights, opens new tab MILAN, April 30 (Reuters) - Italy's antitrust authority said on Thursday it had closed investigations into three AI companies over allegedly unfair commercial practices involving generative artificial intelligence, after accepting binding commitments from them. The regulator, known as the AGCM, also polices consumer rights. It said it had targeted China's DeepSeek, France's Mistral AI SAS and Turkey's Scaleup Yazilim Hizmetleri Anonim Şirketi over risks of so-called AI hallucinations - the generation of
White House AI Memo Hits Issues in Anthropic-Pentagon Feud - Bloomberg
White House AI Memo Hits Issues in Anthropic-Pentagon Feud - Bloomberg Anthropic: BackForward A member of the military specialised in cyber defense works on servers. Photographer: Philippe Huguen/AFP/Getty Images Gift this article Contact us:Provide news feedback or report an error Confidential tip?Send a tip to our reporters Site feedback:Take our Survey By Maggie Eastland, Mackenzie Hawkins, and Hadriana Lowenkron April 30, 2026 at 1:08 PM UTC Updated on April 30, 2026 at 3:32 PM UTC Save Translate White House officials are preparing a wide-ranging artificial intelligence policy memo that outlines requirements for AI deployment by national security agencies, some of which address the issues driving a bitter dispute between the Pentagon and Anthropic PBC over military use of the firm’s technology, according to people familiar with the matter. In the works for months, the
Rep. Dan Goldman on AI companies: 'Regulate them all'
# Rep. Dan Goldman on AI companies: 'Regulate them all' Published: 2026-04-30T18:26:41+00:00 ## Summary Rep. Dan Goldman, who is facing a primary challenge from former New York City comptroller Brad Lander, answered questions about AI and data center regulation, his support for U.S. military aid for Israel, and his understanding of the situation. The Post's Anna Liss-Roy is asking candidates for their candid candidacies regarding these issues. ## Story Rep. Dan Goldman on AI companies: 'Regulate them all' (Anna Liss-Roy/The Washington Post) Up next in Politics Politics # Rep. Dan Goldman on AI companies: 'Regulate them all' April 30, 2026 | 6:16 PM GMT Rep. Dan Goldman (D-New York), who faces a primary challenge from former New York City comptroller Brad Lander, answered rapid-fire questions on AI and data center regulation, his support of U.S. military aid for Israel and what sets
Regulatory Uncertainty Surrounding Cybersecurity Risks in Advanced General-Purpose Models
The lack of standardized regulation for cybersecurity risks in frontier models creates an uneven playing field for developers. As labs self-report risks, government intervention may disproportionately restrict certain firms while others proceed with releases.
Setting AI Rules of Engagement - CEPA
Setting AI Rules of Engagement - CEPA # Setting AI Rules of Engagement Governments are struggling with outdated tactics to assert control over frontier AI models, at home and abroad. By April 30, 2026 It’s a telling contradiction. The US National Security Agency is reportedly using Anthropic’s Mythos model, while the Pentagon has designated the same company as a supply chain risk and banned federal agencies from using its products. Governments are struggling to form a coherent approach to an emerging, powerful technology that they cannot do without, particularly as they see China challenging Western AI leadership. Existing frameworks, from financial regulation to cybersecurity law and AI legislation, don’t seem to fit, as some companies want to put restrictions on how their technology is used. Who should set the terms? Also unanswered is how to meet the largest challenge with AI:
Europe Struggles to Address AI Liability - CEPA
European AI developers might flee to countries with simple, unified rules. A regulatory response should address those risks — fast. If Europe’s governments put in their own rules, it will become a steep challenge to impose a bloc-wide standard. Without credible liability rules, the EU risks an awkward equilibrium: high compliance ...
Get the full executive brief
Receive curated insights with practical implications for strategy, operations, and governance.