Nvidia launches Open Agent Safety Platform, a reference design to stop AI
Nvidia launches the Open Agent Safety Platform, a reference design to stop AI agents from escaping, made up of OpenShell for CPUs and Sentry for network chips
TL;DRNvidia built tools to keep AI agents from breaking free of their constraints.
Why it matters: As AI agents grow more autonomous, containment failures could cause real damage—this addresses a critical safety gap.
Q&A with Cloudflare CEO Matthew Prince on bot traffic potentially
TL;DRCloudflare CEO discusses charging AI bots for web access to combat exploding bot traffic.
NXP and TSMC affiliate Vanguard inaugurate their joint advanced chip
TL;DRNXP and TSMC's partner opened an advanced chip factory in Singapore, ramping production in 2027.
Latest
Hundreds of people attended a pro-AI party in DC on September 26
Pro-AI advocates threw a party in DC to push back against bipartisan skepticism of data centers.
A profile of South Korea Deputy PM Bae Kyung-hoon, an AI evangelist
South Korea's Deputy PM is pushing massive AI investment despite global concerns about risks.
Holo4: powering generalist computer-use agents
New AI model enables computers to perform diverse tasks autonomously.
OpenAI’s Sam Altman Will Not Attend Australia Senate AI Inquiry
Sam Altman won't testify at Australian Senate hearing on AI security breach.
The Dutch government is testing DAWO, a NixOS-based digital work
Dutch government building Microsoft-free work system after ICC access cuts.
Solving Math’s Greatest Problems Was an Art Form. Then Came AI
AI is changing how math gets solved, threatening traditional creative approaches.
Bitget resumes Bitcoin withdrawals after $387.5 million crypto heist
Bitget restored Bitcoin withdrawals after recovering from a $387.5 million hack.
HMRC vowed to break up with Capgemini then paid it another £4.2B
UK tax authority extending £4.2B contract with French firm despite breakup pledge.
Nvidia’s Answer to Rogue Agents Is an Open-Source AI Security System
Nvidia released open-source software to prevent AI agents from breaking free.