New experts join Google’s AI & Economy team
We are expanding our AI & Economy team with world-class academic advisors, fellows, and core internal researchers.
Models blog.google signal 10 primary
Archived ranking
240 rankedSecurity researchers used Claude to help them hack into OpenAI
A team of three independent security researchers at Hacktron says it took less than 72 hours for them to hack into OpenAI employee accounts using Anthropic's Claude Opus 4.8 and 5, the Wall Street Journal reports.
Chips theverge.com signal 9 verified
Researchers used Claude to hack OpenAI
Researchers used Claude to reach an OpenAI employee account and sensitive GitHub data.
Startups arstechnica.com signal 9 verified
A Tsinghua Professor’s Stealth LLM Startup Hits $1.4 Billion Valuation
A secretive Chinese AI model startup founded in February by a Tsinghua University professor is now valued at more than $1.4 billion, after raising $400 million in three funding rounds from investors including Tencent, according to a person with direct knowledge of the matter.
Policy theinformation.com signal 9 verified
Bug Hunters Used Claude to Hack OpenAI
A team of cybersecurity researchers with startup Hacktron AI used Anthropic’s Claude to hack into OpenAI, gaining access to a key repository of OpenAI’s software, the researchers revealed Thursday. The hack, which occurred on July 25, further demonstrates AI companies’ vulnerability to...
Policy theinformation.com signal 9 verified
Cooley Partners with OpenAI on Legal Tool For Drafting SEC Filings
OpenAI and law firm Cooley co-launched a product, called GO Public, that can draft S-1 filings, the document companies must submit to the Securities and Exchange Commission before they can go public, Cooley said Thursday. Lawyers at Cooley can prompt GO Public, which also has access to Cooley...
Policy theinformation.com signal 9 verified
What X is saying
visibilityNo current X pulse in the 24-hour window.
Signals
All →No signals in the last 7 days.
Just in
Story
lesswrong.com
A non-generative model as a trusted monitor for AI Control: Testing TypeSafe's Jev
Story techradar.comUS inherits Venezuelan surveillance state built on banned Chinese tech and AI — and it shows no signs of switching
Story blog.googleEarn continuing education and college credits for AI educator training.
Story techcrunch.comFamily offices are clamoring for AI investments
Topic radar
All →
AI infrastructure and compute
+7.2
Expanding hardware demand is driving semiconductor projections while federal and state policymakers advance legislative measures requiring high-capacity AI data centers to cover electrical grid infrastructure upgrade costs.
Capital and consolidation +2.6Investment capital continues flowing into specialized artificial intelligence firms, with family offices and venture funds backing startups focused on enterprise vulnerability management, manufacturing quality, and open-weight foundation models.
Robotics and embodied AI +1.3Robotics initiatives are advancing through open access to specialized training datasets alongside supplier evaluations and inspections for ongoing robotics manufacturing.
AI policy and governance -12.7Regulators and technology providers are addressing AI safety risks, exploring oversight mechanisms such as emergency controls while evaluating disclosures regarding models evading oversight or altering reasoning processes.
Quick takes
Frontier AI developers and infrastructure providers are turning to grid-finding coalitions and smaller distributed deals to secure data center power as rapid capacity buildouts encounter energy limits.Cybersecurity researchers are utilizing frontier models to exploit software environments while uncovering shared vulnerabilities across major coding agent runtimes, shifting AI safety attention toward direct enterprise security exposure.Frontier model architectures are expanding into real-time speech and extended reasoning, positioning low-latency dialogue as an integrated interface for interactive agent applications.AI frontier labs and hyperscalers are actively securing power and infrastructure through grid-capacity consortiums and distributed data-center contracts to sustain long-term compute expansion.Widespread deployment of autonomous coding agents across multiple vendor ecosystems is introducing shared security vulnerabilities and expanding enterprise attack surfaces.Accelerated silicon roadmaps from domestic hardware manufacturers signal intensifying efforts to establish alternative accelerator supply chains and narrow computing disparities with global market suppliers.Mistral's €3 billion Series D round at a valuation exceeding €21 billion demonstrates significant capital concentration behind sovereign, open-weight foundational models.Escalating infrastructure constraints are driving AI developers toward modular deployments and multi-party energy coalitions targeting up to 100 GW of grid capacity for data centers.
Just in
Story
lesswrong.com
A non-generative model as a trusted monitor for AI Control: Testing TypeSafe's Jev
Story techradar.comUS inherits Venezuelan surveillance state built on banned Chinese tech and AI — and it shows no signs of switching
Story blog.googleEarn continuing education and college credits for AI educator training.
Story techcrunch.comFamily offices are clamoring for AI investments
Market pulse
neutral
AI pulse
48/100
AI-linked equities are under pressure, with Super Micro Computer -3.79%, Oracle -3.33%, Broadcom. +2.33% driving the tracked basket lower.
| Ticker | Company | Move |
|---|---|---|
| SMCI | Super Micro Computer | -3.79% |
| ORCL | Oracle | -3.33% |
| AVGO | Broadcom. | +2.33% |
| META | Meta Platforms | -1.70% |