AI News Archive: May 4, 2026 — Part 2
Sourced from 500+ daily AI sources, scored by relevance.
- Nvidia's exposure to Asian supply chains for components hits 90% of its production costs — marked increase from 65% could intensify as physical AI adds even more exposure
Asian suppliers now represent roughly 90% of Nvidia's production costs, up from about 65% a year earlier, according to data compiled by Bloomberg.
- Google’s AI deal with the Pentagon has sparked employee backlash. But don't expect a repeat of Project Maven
Google’s AI deal with the Pentagon has sparked employee backlash. But don't expect a repeat of Project Maven Fortune
- SAP Moves to Block OpenClaw and Other Unauthorized AI Agents
SAP Moves to Block OpenClaw and Other Unauthorized AI Agents The Information
Score: 60🌐 MovesMay 4, 2026https://www.theinformation.com/articles/sap-moves-block-openclaw-unauthorized-ai-agents - Data centers at sea: Oregon’s Panthalassa nets $140M led by Peter Thiel for wave-powered AI
By combining massive floating power generators with onsite AI hardware, Panthalassa turns the ocean into a self-sustaining computing powerhouse — all without needing a single mile of underwater power cables. Read More
- AI insurance claims startup Reserv raises $125M as it eyes massive expansion
The New York company plans to grow its claims capacity from 500,000 annually to 30 million in four years. It already serves nearly 200 insurers.
Score: 60💰 MoneyMay 4, 2026https://www.bizjournals.com/newyork/news/2026/05/04/reserv-raises-125m.html?ana=brss_6150 - Carlyle buys majority stake in Knack, EqualizeRCM to build AI-led healthcare billing platform
The private equity firm is betting on consolidation in a fragmented US billing services market, with plans to scale via acquisitions and global delivery
- What Anthropic’s Mythos can hack in hours, Indian banks take months to fix
The post What Anthropic’s Mythos can hack in hours, Indian banks take months to fix appeared first on The Ken .
Score: 60🌐 MovesMay 4, 2026https://the-ken.com/story/what-anthropics-mythos-can-hack-in-hours-indian-banks-take-months-to-fix/ - 10-MW nuclear reactor on truck bed: Power bank for AI data centres
10-MW nuclear reactor on truck bed: Power bank for AI data centres Gulf News
- Microsoft takes Agent 365 out of preview as shadow AI becomes an enterprise threat
Microsoft last week took Agent 365 , its management platform for AI agents, out of preview and into general availability — a move that signals the software giant believes the governance challenge around autonomous AI is no longer theoretical but operational and urgent. The product, first announced at Microsoft's Ignite conference in November, positions itself as a unified control plane that lets enterprise IT and security teams observe, govern, and secure AI agents wherever they run: inside Microsoft's own ecosystem, on third-party cloud platforms like AWS Bedrock and Google Cloud , on employee endpoints, and increasingly across a sprawling ecosystem of SaaS agents built by partner software companies. But the most striking element of the launch isn't the general availability milestone itself. It's Microsoft's aggressive push into discovering and managing local AI agents — the coding assistants, personal productivity tools, and autonomous workflows that employees are installing on their
- The global memory shortage: The hidden bottleneck behind the AI boom
As AI demand soars, global memory shortages are driving costs up and reshaping the tech landscape.
Score: 59🌐 MovesMay 4, 2026https://www.techradar.com/pro/the-global-memory-shortage-the-hidden-bottleneck-behind-the-ai-boom - Q&A: Why the threat of Anthropic’s Mythos demonstrates the need for sovereign AI
CIRA VP Jon Ferguson on how the Project Glasswing preview could leave Canadian firms out in the cold. The post Q&A: Why the threat of Anthropic’s Mythos demonstrates the need for sovereign AI first appeared on BetaKit .
Score: 59🌐 MovesMay 4, 2026https://betakit.com/qa-why-the-threat-of-anthropics-mythos-demonstrates-the-need-for-sovereign-ai/ - DNA-reading AI reconstructs ancestry in minutes, matching top statistical methods
Researchers at the University of Oregon have developed an artificial intelligence tool that can read genetic code the way large language models like ChatGPT read text. Scanning the genome for biological mutation patterns, the computer model traces pairs of genes back in time to their last common ancestor.
- Citi launches Arc to scale AI agents across the business
The bank is looking to the technology to automate manual tasks such as research and data analysis.
Score: 59🌐 MovesMay 4, 2026https://www.bankingdive.com/news/citi-launches-arc-scale-ai-agents/819177/ - Trump’s former AI czar says the quiet part out loud on the economy: ‘Stopping progress in AI would be equivalent to halting the US economy’
Trump’s former AI czar says the quiet part out loud on the economy: ‘Stopping progress in AI would be equivalent to halting the US economy’ Fortune
Score: 59🌐 MovesMay 4, 2026https://fortune.com/2026/05/04/trump-ai-czar-david-sacks-american-gdp-economy/ - The Chinese Streaming Industry Is Being Gutted by AI-Generated Shows
"They’re still just imitating humans or trying to make things more humanlike." The post The Chinese Streaming Industry Is Being Gutted by AI-Generated Shows appeared first on Futurism .
Score: 59🌐 MovesMay 4, 2026https://futurism.com/artificial-intelligence/chinese-streaming-industry-gutted-ai-generated-shows - Anthropic and OpenAI are both launching joint ventures for enterprise AI services
Both Anthropic and OpenAI have partnered with asset managers to more aggressively market their enterprise AI products.
- Irretrievability; or, Murphy's Curse of Oneshotness upon ASI
Example 1: The Viking 1 lander In the 1970s, NASA sent a pair of probes to Mars, the Viking 1 and Viking 2 missions. Total cost of $1B (1970), equivalent to about $7B (2025). The Viking 1 probe operated on Mars's surface for six years, before its battery began to seriously degrade. One might have thought a battery problem like that would spell the irrevocable end of the mission. The probe had already launched and was now on Mars, very far away and out of reach of any human technician's fixing fingers. Was it not inevitable, then, that if any kind of technical problem were to be discovered long after the space launch in August 1975, nothing could possibly be done? But the foresightful engineers of the Viking 1 probe had devised a plan for just this class of eventuality, which they had foreseen in general, if not in exact specifics. They had built the Viking 1 probe to accept software updates by radio receiver, transmitted from Earth. On November 11, 1982, Earth sent an update to the Vik
Score: 58🌐 MovesMay 4, 2026https://www.lesswrong.com/posts/fbrz9xhKpEeTKw5zL/irretrievability-or-murphy-s-curse-of-oneshotness-upon-asi - CISA Weighs Cutting Deadlines to Fix Digital Flaws Amid Worries Over AI
U.S. cybersecurity officials are considering sharply shorter deadlines for fixing critical flaws in government IT systems, amid concerns hackers could exploit them using artificial‑intelligence tools such as Anthropic’s Mythos, people familiar with the matter said. The move, which has not …
- UK Government Announces Plans to Grow National AI Infrastructure
The UK Government is to support the development of Britain’s AI hardware infrastructure, while also committing to work in establishing international standards for the deployment of AI. The post UK Government Announces Plans to Grow National AI Infrastructure appeared first on TechRepublic .
Score: 58🌐 MovesMay 4, 2026https://www.techrepublic.com/article/uk-government-announces-plans-to-grow-national-ai-infrastructure/ - The heartland’s revenge: how AI is reindustrializing the American interior
AI investment is increasingly driving construction growth in the U.S. The post The heartland’s revenge: how AI is reindustrializing the American interior appeared first on FreightWaves .
Score: 58🌐 MovesMay 4, 2026https://www.freightwaves.com/news/the-heartlands-revenge-how-ai-is-reindustrializing-the-american-interior - Meta Cut 8,000 Jobs to Fund AI. Zuckerberg Said It Out Loud. Nobody Noticed.
Meta and Microsoft cut 23,000 jobs and spent $700 billion on AI in the same quarter. Continue reading on Towards AI »
- Sam Altman sets out OpenAI's three key focus areas for next phase of growth
OpenAI CEO Sam Altman reveals future plans. The company will focus on accelerating science and boosting economic productivity. Building personal AGI is also a key goal.
- Nvidia Backs DeepInfra in $107 Million Raise
Cloud inference platform DeepInfra closes a $107 million in Series B funding round, backed by the likes of Nvidia and Samsung, as it aims to tackle bottlenecks in AI compute. DeepInfra CEO Nikola Borisov speaks with Caroline Hyde on “Bloomberg Tech.” (Source: Bloomberg)
Score: 58💰 MoneyMay 4, 2026https://www.bloomberg.com/news/videos/2026-05-04/nvidia-backs-deepinfra-in-107-million-raise-video - Palantir posts its fastest revenue growth ever while calling out ‘AI slop’
Palantir posts its fastest revenue growth ever while calling out ‘AI slop’
- From data to diagnosis: How AI is helping doctors detect disease
From data to diagnosis: How AI is helping doctors detect disease Gulf News
Score: 58🌐 MovesMay 4, 2026https://gulfnews.com/uae/health/from-data-to-diagnosis-how-ai-is-helping-doctors-detect-disease-1.500528107 - OpenAI says human attention is the bottleneck, so it built a system to let agents manage themselves
OpenAI's new Symphony spec flips the AI coding workflow on its head: instead of developers babysitting multiple Codex sessions, agents pull their own tickets straight from Linear and run until the job is done. The article OpenAI says human attention is the bottleneck, so it built a system to let agents manage themselves appeared first on The Decoder .
- GUC and Wiwynn Collaborate on Silicon-to-System Infrastructure for Next-Gen Hyperscale AI
Collaboration unites SoC design, optical I/O, and liquid-cooled rack integration to help hyperscalers accelerate AI infrastructure. The post GUC and Wiwynn Collaborate on Silicon-to-System Infrastructure for Next-Gen Hyperscale AI appeared first on EE Times .
- The AI electric grid calls for a mixed‑fleet transmission strategy
In the AI-era planning environment, hybridization is not a compromise — it is the engineering approach best aligned with real world system needs, timelines and locational realities, writes NYISO planning engineer Anees Jeddy.
Score: 57🌐 MovesMay 4, 2026https://www.utilitydive.com/news/the-ai-electric-grid-calls-for-a-mixedfleet-transmission-strategy/819116/ - Trolling, memes and deepfakes: How AI is thickening the fog of war
Trolling, memes and deepfakes: How AI is thickening the fog of war reutersinstitute.politics.ox.ac.uk
Score: 57🌐 MovesMay 4, 2026https://reutersinstitute.politics.ox.ac.uk/news/trolling-memes-and-deepfakes-how-ai-thickening-fog-war - Meta looks to space solar to solve AI energy crunch
Meta looks to space solar to solve AI energy crunch YourStory.com
- Antimatter Launches as the World's First Vertically Integrated Neocloud for AI Inference, Plans to Establish Global Headquarters in Hong Kong
Antimatter Launches as the World's First Vertically Integrated Neocloud for AI Inference, Plans to Establish Global Headquarters in Hong Kong The Straits Times
- A bill that bans kids from using AI chatbots is gaining momentum
A bill that bans kids from using AI chatbots is gaining momentum marketplace.org
- Inside AMEX’s agentic commerce stack: How intent contracts and single-use tokens enforce AI transactions
American Express (Amex) is building a system that lets AI agents shop and pay on behalf of users — but right now it’s only within its own payment network, and still involves a black box that could hinder trust and auditability. Amex already participates in agentic commerce protocol projects, especially Google’s Agent Pay Protocol (AP2) , which focuses on interoperability. Amex’s Agentic Commerce Experiences (ACE) developer kit, on the other hand, touches on something most protocols currently lack: Full transaction control in the payment layer. But it still isn't completely transparent in how it handles validation. ACE uses a closed-loop system — serving as both the card issuer and the payment network — to validate agent-led transactions. Luke Gebb, Amex's EVP and global head of innovation, told VentureBeat that the company believes this model is the missing piece in agentic commerce. “Some of what is missing so far is the perspective of a company like ours: We feel that trust and secur
- The Royal Navy will get its first large unmanned warship in just two years. Really?
The Royal Navy will get its first large unmanned warship in just two years. Really? The Telegraph
Score: 56🌐 MovesMay 4, 2026https://www.telegraph.co.uk/news/2026/05/04/royal-navy-large-unmanned-surface-vessel-hybrid-nato/ - AI energy reckoning renews nuclear debate 40 years after Chernobyl
AI energy reckoning renews nuclear debate 40 years after Chernobyl The National
Score: 56🌐 MovesMay 4, 2026https://www.thenationalnews.com/future/technology/2026/05/04/chernobyl-ai-nuclear-support-three-mile-island/ - ServiceNow Sees $30 Billion Revenue by 2030 on AI Uplift
ServiceNow Inc. projected it would generate $30 billion of subscription revenue in 2030, attributing the strong outlook to traction from its AI products.
- Micron CEO warns 'AI is in very early innings' and it will 'need more memory' — another ominous sign the RAM crisis isn't going anywhere
Micron has warned on the gravity of the RAM crisis, meaning that every major memory chip maker has now done so.
- Biopharma giant Sanofi to grow Toronto presence with expanded AI centre
French company says $294-million investment will create 50 new jobs. The post Biopharma giant Sanofi to grow Toronto presence with expanded AI centre first appeared on BetaKit .
Score: 56🌐 MovesMay 4, 2026https://betakit.com/biopharma-giant-sanofi-to-grow-toronto-presence-with-expanded-ai-centre/ - Why some AI tools are being banned by the U.S. government — and what it means for you
Why some AI tools are being banned by the U.S. government — and what it means for you Tom's Guide
- KAIST: AI Learns to Say “I’m Not Sure” … Reducing Overconfidence and Improving Reliability
KAIST: AI Learns to Say “I’m Not Sure” … Reducing Overconfidence and Improving Reliability EurekAlert!
- Nvidia explains why sovereign AI is important
Nvidia explains why sovereign AI is important The National
Score: 56🌐 MovesMay 4, 2026https://www.thenationalnews.com/video/Km3u6gaS/nvidia-explains-why-sovereign-ai-is-important/ - Google, VNG launch Applied AI Lab to accelerate Vietnam’s AI capabilities
Google, VNG launch Applied AI Lab to accelerate Vietnam’s AI capabilities DealStreetAsia
- Senators seek Labor-led database on AI workforce impacts
The Workforce Transparency Act from Sens. Mark Warner and Ted Budd would charge the DOL with creating a public resource with “aggregated workforce transparency data.” The post Senators seek Labor-led database on AI workforce impacts appeared first on FedScoop .
Score: 55🌐 MovesMay 4, 2026https://fedscoop.com/labor-department-ai-workforce-data-warner-budd-bill/ - The Silicon Protocol: How to Cut LLM Context Costs 80% in Healthcare, Government & Finance (2026)
The Silicon Protocol: How to Optimize LLM Context Windows Without Breaking Production Systems in Healthcare, Finance & Government (2026 Guide) 200K tokens cost $47 per request. Your model stopped paying attention at 80K. The bill arrives anyway. Three context window management patterns. Full injection hits 2x surcharge, costs $47K/month, model loses 30% accuracy past 80K tokens. Selective RAG maintains quality at 32K tokens, costs $7.2K/month. RULER benchmarks show why more context degrades output. Context window costs are the silent budget killer in production LLM systems, accounting for 60–80% of total API spend while delivering diminishing returns past 50K tokens. When organizations deploy large language models with 200K+ token context windows, they assume bigger windows mean better outputs — but RULER benchmark testing shows models lose 30%+ accuracy on mid-context retrieval, pricing surcharges kick in above 200K tokens (2x input cost for Claude, Gemini), and most prompts could run
- OpenAI Opens ChatGPT to OpenClaw's 3.2M Users While Anthropic Blocks Access
OpenAI opens ChatGPT to OpenClaw's 3.2M users, while Anthropic blocks access.
Score: 55🌐 MovesMay 4, 2026https://opentools.ai/news/openai-opens-chatgpt-to-openclaw-3-million-users-anthropic-blocks - Microsoft and OpenAI Rework Alliance, Loosening Exclusive Ties
Microsoft and OpenAI have adjusted the terms of their high-profile partnership, signaling a shift in how the two companies will collaborate as competition in the AI market intensifies.
- Google Intros New Gemini Enterprise Agent Platform
Google Cloud has announced a new platform for building and managing enterprise AI agents, as the company seeks to turn its Gemini models and Vertex AI tooling into a broader system for automating business workflows.
Score: 55🌐 MovesMay 4, 2026https://campustechnology.com/articles/2026/05/04/google-intros-new-gemini-enterprise-agent-platform.aspx - Anthropic targets vulnerability detection gains with Claude Security public beta — here's what users can expect
Anthropic targets vulnerability detection gains with Claude Security public beta — here's what users can expect IT Pro
- AI Deflation is Beginning to Hollow Out India’s IT Revenue Engine
Artificial intelligence, by making the work faster and cheaper, is compressing contract values even before they are signed.
Score: 55🌐 MovesMay 4, 2026https://analyticsindiamag.com/it-services/ai-deflation-is-beginning-to-hollow-out-indias-it-revenue-engine - Japanese scientists push for AI use in medical research and diagnoses
Japanese scientists push for AI use in medical research and diagnoses The Japan Times
Score: 55🌐 MovesMay 4, 2026https://www.japantimes.co.jp/news/2026/05/04/japan/science-health/ai-medicine/