Sat 23 May 2026
Daily Brief — Curated and contextualised by Best Practice AI
DeepSeek Discounts, Starbucks Scraps AI, and Standard Chartered Faces Backlash
TL;DR DeepSeek has permanently reduced prices on its flagship AI model by 75% to boost developer adoption. Starbucks has discontinued its AI inventory tool across North America due to inefficiencies. Standard Chartered's CEO apologized after calling employees losing jobs to AI 'lower-value human capital.' Meanwhile, the US considers tariffs on imported semiconductors to encourage domestic production.
The stories that matter most
Selected and contextualised by the Best Practice AI team
DeepSeek To Make Permanent 75% Discount on Flagship AI Model
DeepSeek said it will make permanent a steep discount on its flagship V4‑Pro model, maintaining prices for developers at a quarter of their original level.
The Fate of AI Depends on Physical Infrastructure, Not Just Algorithms - Bloomberg
The need for compute makes strange bedfellows. On May 6, just a few months after Elon Musk called Anthropic PBC “misanthropic and evil,” he agreed to lease the entire capacity of SpaceX’s Memphis data center to the artificial intelligence firm. Anthropic is now paying $1.25 billion per ...
Exclusive: Starbucks scraps AI inventory tool across North America | Reuters
Starbucks rapidly rolled the tool out to North American stores in September, the company announced at the time. The AI -powered app aimed to replace hand counts of some products with automated ones that were expected to be faster and more accurate.
Workday wants AI to punch in instead of having to hire new recruits
CEO eyes margin gains by keeping headcount flat – bold for a company selling HR software to employers
US Weighs Chip Tariffs to Spur Domestic Growth, Trade Chief Says
The Trump administration continues to weigh US tariffs on imported semiconductors to boost domestic chip manufacturing, though there are no immediate plans to impose any new levies, US Trade Representative Jamieson Greer said.
AI datacenter boom collides with US grid reality
Wood Mackenzie analysts say bit barn operators are in a tough spot
How VCs and founders use inflated ‘ARR’ to crown AI startups | TechCrunch
Some AI startups are stretching traditional revenue metrics when talking about progress publicly. And their investors are fully aware.
Presien Reduces Critical Safety Events on Construction Sites by 70%+ with Claude
Presien utilizes Claude for real-time construction site safety monitoring, demonstrating how Generative AI can be applied to physical operations and risk reduction.
Economics & Markets
Zoom’s Anthropic Investment Has Netted the Company $1 Billion
Zoom Communications Inc., the videoconferencing company, has netted about $1 billion on an investment it made in artificial intelligence startup Anthropic PBC in early 2023.
Exclusive: Modal Labs valued at $4.65 billion as AI coding takes off | Reuters
The startup helps AI companies access the chips they need to run AI tools, called inference.
Sohn 2026: AI Dominates, But Winners Are Elusive | StartupHub.ai
AI dominated the Sohn Investment Conference 2026, but identifying specific winners proved difficult amidst infrastructure and power plays.
3 Years of the AI Stock Market Boom in Charts | Morningstar
A look at how exploding AI revenues have driven returns, lifted non-tech stocks, added to market concentration, and sent unicorn valuations flying.
Earnings Season vs. AI - Simply Wall St
30-year yields at 2007 highs. SpaceX filing at a $2 trillion valuation. SaaS stocks getting torched while their income statements paint a different story. We recap a noisy week, dig into the Q1 numbers behind the headlines, and lay out how to tell an AI value trap from a genuine opportunity.
The big AI shock: Have Indian IT sector stocks lost their lustre? - The Times of India
Globally technology sector stocks are leading stock market rallies. In India, the picture is the opposite.
Silicon Shake-Up: The AI Trade Is Moving Beyond Nvidia | Investing.com
Market Analysis by covering: Intel Corporation, Advanced Micro Devices Inc. Read 's Market Analysis on Investing.com
Labor, Society & Culture
Classical music has survived for centuries. Will AI kill it?
Composers have always experimented with new technology — but the latest advances threaten ‘skill death’ in this centuries-old art form
‘You can’t control everything’: the rise in plastic surgeons asked to create ‘AI face’
Growing numbers of people are seeking improbable cosmetic surgery based on chatbots’ recommendations Plastic surgeons are increasingly concerned about the rise of “AI face”, as more and more clients arrive in their offices with unrealistic AI-generated visions of what they want to look like. Dr Nora Nugent, a cosmetic surgeon from Tunbridge Wells, has seen this first hand. Clients have started coming to her office with photos of themselves beautified by AI and a false expectation that those results are achievable with surgery. She is also the president of the British Association of Aesthetic Plastic Surgeons, and says many colleagues are having similar experiences. Continue reading...
Bill Winters ‘Lower-Value Human’ Apology Not Enough for Unions
Bill Winters’ efforts to restore calm after making some controversial comments on AI have failed to reassure labor organizations, according to one of the world’s largest union federations.
AI Won't Crush the Labor Market, JPMorgan Exec Says - Business Insider
AI doomers say the tech could crush the job market, but a growing chorus of market pros says that it'll reshuffle, not shrink, employment.
How Much Will AI Impact Tomorrow’s Workforce? New Data on the Future of Work with AI | MIT Initiative on the Digital Economy - BrianHeger.com
A new synthesis from MIT IDE research scientists draws on three working papers on different aspects of the impact of AI. Links to all three papers are also provided.
Zuck defends monitoring employees to win AI race in purported leaked audio
Limping Llama model needs a crutch made of surveillance tools
Is a college degree is still worth it? Here are 3 things it can teach you that AI can’t do
College can help safeguard employees from having their jobs offshored to India or the Philippines, Carl Benedikt Frey told Fortune.
There’s Never Been a Better Time to Study Computer Science
Even as AI progresses, coders aren’t doomed.
AI forces rethink of talent as skills gap widens | Cyprus Mail
Artificial intelligence is no longer only changing the tools people use at work, but it is also reshaping what companies, universities and training centres understand by talent. Indeed, speakers at a Cyprus Seeds panel discussion that took place at the Doers Summit in Limassol this week argued ...
Why coding, AI, and real-world projects are the new foundation of B. Tech education - The Hindu
Explore how coding, AI, and real-world projects are transforming B. Tech education and shaping industry-ready engineers in India.
Council Post: How Leaders Can Assist With Upskilling And Reskilling Amid AI Disruption
Upskilling and reskilling requires a coordinated effort across institutions, employers and educators.
Sundar Pichai Understands Why People Are Anxious About A.I.
After a busy Google I/O, the company’s chief executive sits down with the hosts of “Hard Fork” to discuss the future of Google Search, how he’s using A.I. agents and his advice for college graduates.
Why College Students Are Booing AI
The sound of a cosmic howl
Technology & Infrastructure
Autonomous AI Agents: Complete Enterprise Guide 2026
Here’s the uncomfortable truth most enterprise leaders already know but rarely say out loud: the way work gets done inside large organizations hasn’t fundamentally changed. What’s changed is the volume. More data, more systems, more decisions, but the same human-dependent workflows underneath it all. Skilled people are spending their days on tasks that shouldn’t require their skills in the first place. Traditional automation gave us speed on simple, repetitive work. AI ...
Agentic AI Adoption Statistics for 2026 – First Page Sage
Key agentic AI adoption statistics for 2026, including rates by industry, company size, and implementation stage. Includes free PDF download.
ManageEngine rolls out autonomous AI agents across suite
The move gives IT teams autonomous agents for service desks, security and endpoint work, while ManageEngine says customer data stays private.
Your AI agents need a terminal, not just a vector database
When agentic workflows fail, developers often assume the problem lies in the underlying model’s reasoning abilities. In reality, the limited information provided by the retrieval interface is often the primary limiting factor. Researchers at multiple universities propose a technique called direct corpus interaction (DCI) that lets agents bypass embedding models entirely, searching raw corpora directly using standard command-line tools. The limits of classic retrieval In classic retrieval systems such as RAG, documents are chunked, converted into vector representations (or embeddings), and indexed offline in a vector database. When an AI system processes a query, a retriever filters the entire database to return a ranked "top-k" list of document snippets that match the query. All evidence must pass through this scoring mechanism before any downstream reasoning occurs. But modern agentic applications demand much more. "Dense retrieval is very useful for broad semantic recall, but when an agent has to solve a multi-step task, it often needs to search for exact strings, numbers, versions, error codes, file paths, or sparse combinations of clues," the authors of the DCI paper said in comments provided to VentureBeat. "These long-tail details are precisely where semantic similarity can be brittle." Unlike static search, agents must also revise their search plans dynamically after observing partial or localized evidence. Exact lexical constraints and multi-step hypothesis refinement are difficult to execute with semantic retrievers. Because the retriever compresses access into a single step, any critical evidence filtered out by the similarity search cannot be recovered later, no matter how advanced the agent's downstream reasoning capabilities are. As the authors explain, current retrieval pipelines can become a bottleneck because "they decide too early what the agent is allowed to see." Direct corpus interaction This direct access addresses a core problem in enterprise environments: data staleness. Embedding indexes are always a snapshot of a specific moment in time, taking considerable compute and time to build and maintain. "In many enterprise settings, the data is not a stable document collection. It is daily financial reports, live logs, tickets, code commits, configuration files, incident timelines, and internal documents that keep changing," the authors said. DCI lets the agent reason over the current state of the workspace rather than yesterday's vector index. The agent operates in a terminal-like environment where its observations are raw tool outputs such as file paths, matched text spans, and surrounding lines. The core tools provided by DCI are few but highly expressive. Agents use commands like “find” and “glob” to navigate directory structures and locate files. For exact matching, they use “grep” and “rg” to locate specific keywords, regex patterns, and exact strings. When local inspection is needed, tools like “head,” “tail,” “sed,” “cat,” and lightweight Python scripts allow the agent to peek at the context surrounding a match or read specific file sections. The agent can combine these tools via shell pipelines to execute complex search logic in a single step. An agent can pipe commands to enforce strict lexical constraints, such as searching a file for one term and piping the output to search for a second term. It can combine multiple weak clues across a corpus by finding a specific file type, searching for a keyword like "report," and filtering for a year like "2024." It can also immediately verify a hypothesis by inspecting the exact lines around a keyword match. DCI delegates semantic interpretation directly to the agent instead of relying on embedding-based similarity search. The agent can formulate hypotheses, test exact lexical patterns, and extract detailed information that a traditional semantic retriever might miss. The researchers propose two versions of this system. DCI-Agent-Lite is designed as a lightweight, low-cost setup built on the GPT-5.4 nano model and restricted purely to raw terminal interactions like bash commands and basic file reads. Because reading raw files can quickly fill up a smaller model's memory, this version relies on lightweight runtime context-management strategies to sustain long-horizon exploration. DCI-Agent-CC is the higher-performance version, designed for teams with more compute budget. It runs on Claude Code powered by Claude Sonnet 4.6. Claude Code provides stronger prompting, more robust tool orchestration, and superior built-in context handling, which improves the agent's stability during complex, multi-step searches across heterogeneous datasets. DCI in action The researchers tested both versions of DCI across agentic search benchmarks like BrowseComp-Plus, knowledge-intensive QA with single-hop and multi-hop reasoning, and information retrieval ranking in tasks requiring domain-specific reasoning and scientific fact-checking. They tested DCI against three baselines. The first included open-weight retrieval agents such as Search-R1 and proprietary agents powered by frontier models like GPT-5 and Claude Sonnet 4.6, paired with standard retrievers. The second baseline included classical sparse retrievers like BM25 and dense retrievers like OpenAI's text-embedding-3-large and Qwen3-Embedding-8B. The third baseline consisted of high-performing reasoning-oriented re-rankers like ReasonRank-32B and Rank-R1. DCI systematically outperformed the baselines, according to the researchers. On the complex BrowseComp-Plus benchmark, swapping a traditional Qwen3 semantic retriever for DCI on a Claude Sonnet 4.6 backbone improved accuracy from 69.0% to 80.0% while reducing the API cost from $1,440 to $1,016. The return on investment for lightweight agents was also noticeable. DCI-Agent-Lite with GPT-5.4 nano competed with the OpenAI o3 model using traditional retrieval while cutting costs by more than $600. On multi-hop QA benchmarks, DCI-Agent-CC reached an 83.0% average accuracy, improving on the strongest open-weight retrieval baseline by 30.7 points, according to the researchers. The data shows that DCI has lower overall document recall than dense embedding models, but once it finds a relevant document, it extracts substantially more value from it. "If an enterprise AI lead asked where DCI is most clearly useful, I would point to tasks that require exact evidence localization in a dynamic workspace: debugging production incidents, searching large codebases, analyzing logs, compliance investigation, audit trails, or multi-document root-cause analysis," the researchers note. In one complex deep-research task, the agent had to identify a specific soccer match based on 12 interlocking clues, including exact attendance, yellow cards, and player birth dates. A traditional retriever would fail by surfacing short, disconnected snippets. Instead, the DCI agent explored the file directory, read specific lines of a 1990 England versus Belgium match report to verify the exact number of substitutions, pulled a specific quote from an interview file, and verified the exact birth dates of two players by peeking into their Wikipedia text files. By chaining these simple commands, DCI ensures that no evidence is permanently lost behind a flawed semantic search algorithm. Limits and practical implementation of DCI DCI has a clear operating envelope where it scales excellently in search depth but struggles with search breadth. When the experimental corpus was expanded from 100,000 to 400,000 documents, the system's accuracy dropped significantly and the average number of tool calls rose. While DCI is powerful once a promising document is found, the cost of locating that initial useful anchor document grows sharply as the size of the candidate space increases. DCI also has lower broad document recall compared to dense embedding models. It trades exhaustive recall for high-resolution, local precision. If an enterprise workflow strictly requires finding every single relevant document across a massive dataset, DCI may not be the right tool. Granting an agent expressive tools like an unrestricted bash shell increases latency and compute costs due to the high volume of iterative tool calls required to complete a search. It also creates significant context-management and security challenges for IT departments. "Tool calls can return large outputs; long trajectories can fill the context window; and raw terminal access requires sandboxing, permission control, and careful engineering," the authors said. To manage the context window, the researchers found that moderate truncation and compaction help the agent sustain longer searches, whereas overly aggressive summarization tends to discard useful evidence. Because of these operational realities, DCI is not meant to be a mandatory replacement for existing vector infrastructure. Instead, it serves as a complementary one. "For orchestration engineers and data architects, our view is that the most practical near-term deployment pattern is hybrid," the authors said. Semantic retrieval can still provide high-recall candidate discovery when a user's intent is broad or underspecified. "DCI can then operate as a precision and verification layer: the agent can search within the retrieved documents, expand from them into neighboring files, check exact constraints, and combine weak signals across documents." The researchers have released the code for DCI under the permissive MIT license. "Longer term, DCI changes how we think about enterprise data. Data will not only need to be stored for humans or indexed for search engines; it will need to be organized for agents that can inspect, compare, grep, trace, and verify," the authors conclude. "File names, timestamps, stable identifiers, metadata, version history, and machine-readable structure become part of the retrieval interface."
Council Post: The Real Barrier To Enterprise AI Isn’t Capability; It’s Control
How do organizations allow AI systems to operate autonomously without introducing unacceptable levels of risk?
The Fate of AI Depends on Physical Infrastructure, Not Just Algorithms - Bloomberg
The need for compute makes strange bedfellows. On May 6, just a few months after Elon Musk called Anthropic PBC “misanthropic and evil,” he agreed to lease the entire capacity of SpaceX’s Memphis data center to the artificial intelligence firm. Anthropic is now paying $1.25 billion per ...
Bloom Energy Powers the AI Revolution With 130% Revenue Surge and $5B Brookfield Deal - Blockonomi
Bloom projects 30% of all data center sites will rely on onsite power as a primary energy source by 2030. Bloom Energy is gaining ground as one of the most closely watched names in AI infrastructure.
Capital Is Not the Bottleneck in AI Infrastructure. Here Is What Is.
Asia-Pacific is at $180 billion-plus and accelerating. The constraint story varies by sub-region. Singapore is capacity-constrained, commanding premium rents. India is solving a sovereign AI policy constraint Google’s $10 billion commitment is not about returns; it is about a government telling a hyperscaler that local compute ...
Yeebo Ramps Up AI Computing Expansion with Subsidiary Suanova’s TaaS Rollout at Cyberport
The Company's core business spans ... consumer applications. Headquartered in Hong Kong, Yeebo operates its manufacturing operations primarily in the Guangdong and Jiangsu provinces, supporting a global sales network that ensures localized service and support for its international clientele. In alignment with its long-term strategic vision, Yeebo is leveraging its robust operational foundation to expand into the Artificial Intelligence ("AI") compute and related sectors...
The new arms race in computing power
Military capability depends increasingly on data centres. Now governments outpaced in AI are looking to experimental technologies
AI datacenter boom collides with US grid reality
Wood Mackenzie analysts say bit barn operators are in a tough spot
AI Revolution - Infrastructure Struggles to Keep Up with Demand | VoIP Review
A recent study by HBR Analytic Services highlights a challenge facing businesses as they strive to implement agentic AI solutions.
AI Inference Pulls Infrastructure Back Into Metro Data Centers
Shane Snider is Senior News Writer ... energy systems driving modern compute expansion. His reporting focuses on the operational, economic, and environmental forces reshaping digital infrastructure, including AI factories, utility constraints, liquid cooling, renewable energy ...
Inside the Energy Challenges Facing AI Data Centres | Data Centre Magazine
At Data Centre LIVE, Centrica’s Director of Research & Innovation, Dr Ben Krikler, explored whether AI is the grid’s biggest threat, or its smartest fix
The Gulf’s AI Boom Has an Undersea Cable Problem | WIRED
Hyperscalers are pushing the Gulf to rethink internet infrastructure as AI raises the stakes of cable disruptions.
SpaceX US securities filing reveals xAI details, insight into AI regulatory risks
The S-1 filing by SpaceX provides significant insight into the legal and regulatory risks surrounding xAI's Grok chatbot and plans for "Orbital AI compute" data centers.
Nvidia CEO Urges Super Micro to Tighten Up Amid Taiwan Crackdown
Nvidia Corp. Chief Executive Officer Jensen Huang urged Super Micro Computer Inc. to tighten up on compliance after Taiwan detained three people this week for allegedly making fraudulent declarations about AI servers made by its US partner.
Minor edits to AI skills can make agents go rogue
Text is the new attack
Dirty Frag, Copy Fail, Fragnesia: The start of a worrisome Linux security trend
Or is it just life today, with AI constantly digging through code repositories in search of security holes?
The Growing Cybersecurity Risks To The Supply Chain In The AI Era
Supply chains are a primary target for cybercriminals and provide the foundation of global commerce in the hyper-connected digital ecosystem of today
Adoption, Deployment & Impact
Exclusive: Starbucks scraps AI inventory tool across North America | Reuters
Starbucks rapidly rolled the tool out to North American stores in September, the company announced at the time. The AI -powered app aimed to replace hand counts of some products with automated ones that were expected to be faster and more accurate.
When AI Meets Reality: Why Responsible AI Adoption Begins with Data and Governance | by Data & Policy Blog | May, 2026 | Medium
When AI Meets Reality: Why Responsible AI Adoption Begins with Data and Governance By Anastasija Nikiforova Artificial Intelligence (AI) is often presented as the defining technology of our era — …
In Today’s AI World, Infrastructure Proactive Planning is Critical.
That means making a deliberate ... on infrastructure you rent elastically. On-premises remains the right home for workloads with consistent, predictable demand profiles; data that carries sovereignty or compliance constraints; latency-sensitive processing that cannot tolerate a network hop; and compute that you've already invested in and is running efficiently. Cloud - specifically Azure - earns its place as the elastic layer: the capacity you reach ...
Singapore launches AI playbook to steer enterprise transformation
Singapore has launched an AI for Enterprise Impact Playbook to guide businesses through AI adoption, workforce upskilling, and strategic implementation.
The hidden flaw in insurance AI adoption for advisors and carriers - Insurance News | InsuranceNewsNet
Many insurers are still using AI for existing underwriting and claims rather than redesigning how those workflows operate.
The Promise of AI Is Not Enough - Leadership Is
What Chen, Chen, & Lin (2020) documented in their review of AI applications in education holds here: outcomes vary dramatically based not on the presence of the technology, but on how intentionally it is integrated into learning. Personalized tutoring systems produce results when embedded in thoughtful instructional design.
Artificial Intelligence: KDEM CEO Highlights Key Barriers to AI Implementation for Enterprises, ETTelecom
Artificial Intelligence: Enterprises must overcome data access, skill shortages, and knowledge gaps to successfully harness AI technology and achieve business value, according to KDEM CEO Sanjeev Kumar Gupta. Explore the insights from industry leaders on navigating the AI landscape.
How AI Can Help Emerging-Market Startups Build the Systems They’re Missing
AI adoption in emerging markets will not be won by the startups with the flashiest demos.
Stellantis, Qualcomm Expand Partnership for Vehicle Tech
Qualcomm's head of Automotive, Robotics and IoT Businesses Nakul Duggal said that the new vehicle technology partnership the semiconductor company announced with Stellantis will allow the automaker to modernize technology across the board and move fast as the automation and self-driving landscape grows. Duggal said that the automotive industry is moving quickly and he sees more intelligent cars becoming out over the next few years. (Source: Bloomberg)
AI may speed up search for drugs to treat brain conditions
Researchers hope the work will help identify affordable, effective drugs to treat conditions like MND.
Waymo pauses robotaxis in five US cities after cars drive into flooded roads
A Waymo spokesperson said it had expanded the temporary pause "out of an abundance of caution".
Final frontier for meds? UK startup sends drug-making into space
BioOrbit hopes drug-crystallisation technology will lead to self-injected cancer treatment that could save millions Onboard a SpaceX flight last week was a remarkable piece of cargo – a hi-tech box destined for the International Space Station to grow ultra-pure protein crystals, with the aim of producing self-injected cancer drugs. A British startup, BioOrbit, has developed the drug-crystallisation technology at its labs in London and launched Box-E, a compact unit the size of a microwave, on the 15 May rocket from Kennedy Space Center in Florida. Continue reading...
Geopolitics, Policy & Governance
The new Luddite movement
If governments don’t slow AI down, voters — like their predecessors — might take matters into their own hands
Gavin Newsom Warns 'The System Is Broken' As California Launches Sweeping AI Worker Protection Plan Amid - Benzinga
California governor Gavin Newsom unveiled a major AI worker protection plan amid rising automation fears and tech layoffs.
US deepfake legislation would expand safe harbor, takedown system
A revised version of the bipartisan NO FAKES Act aims to establish personal property rights for digital likenesses while expanding safe harbor protections and notice-and-takedown systems.
Bill regulating powerful AI models advances as advocates say it’s only the first step | NPR Illinois
The measure aims to increase transparency and safety protocols around ‘frontier’ AI
AI Communication Crisis Turns Tech Into Political Liability | Opinion - Newsweek
Big Tech is investing billions into the future and almost nothing into explaining it.
FTC Investigates Arm Over Its Licensing Amid AI Chip Boom | Cloud News
Arm Holdings has gone from being the quiet provider of the architecture powering billions of smartphones to becoming an increasingly influential player in the
Governance Concerns Over SpaceX IPO & the Upside of Financing AI
New York City Comptroller Mark Levine discusses the governance concerns over the SpaceX IPO. He talks about the potential legislative changes that could be made to curb these mega-IPOs. He speaks with Romaine Bostick on “The Close.” (Source: Bloomberg)
Get the full executive brief
Receive curated insights with practical implications for strategy, operations, and governance.