Personal AI agents are increasingly being released and used to browse and negotiate on the web, but many websites are blocking them and platforms may redesign the internet around agent traffic.
Agentic web traffic tripled from August to September.
As a result, customer-facing websites and verification flows could become harder for human browsing while rules evolve over whether agents can access sites and bargain for better deals.
Anthropic is creating an Anthropic Public Benefit Fund and committing 5% of its shares to it. The fund could grow as the company’s shares rise, and the company’s post-addition pledge means about 20% of shares could be dedicated to the public good. This expands how Anthropic plans to fund its mission by enabling larger donations to traditional nonprofits or targeted distributions and follows its $200 million AI labor market study.
A speaker at the IMF’s meeting in Bangkok argued that AI systems are starting to change scientific progress by operating at the frontier of human research, sometimes autonomously. They cited test results such as Claude Opus 5.5 rising from 8% at Humanity’s Last Exam’s January 2025 launch to 67.7% “as of this week,” alongside similar steep gains on other science-oriented benchmarks. The takeaway is that policymakers should plan for faster overall scientific and economic change and manage corresponding risks in how AI is developed and deployed.
Illinois Governor JB Pritzker said AI and other tech forces should lead government to take a more active, tougher stance on power, privacy, data centers, and employment impacts rather than defer to big companies. He warned that unemployment could rise from about 4% to 7% and possibly 10% as technology disrupts jobs. As a result, he argued states should push back on corporate demands while preparing worker protections instead of relying mainly on incentives to attract data centers and other AI-linked infrastructure.
China’s commerce and science ministries proposed adding 23 AI technologies, including large model weights, to the country’s export control list. The draft would apply to large models with at least 2 trillion parameters or in the world’s top 10 on tests. A temporary export-registration rule began 16 October and covered exports would require registration and licenses, while open-source code and published research are exempt and overseas access via apps may continue under certain conditions.
US agencies said DeepSeek and Alibaba used distillation at industrial scale, comparing it to copying homework rather than doing original work. The advisory was issued jointly by the NSA, FBI, and CISA. The dispute escalates with Beijing pushing back and hinting at retaliation.
Democrats are signaling they would increase scrutiny of AI companies if they regain committee leadership after the November midterms. The article points to a plan to press the industry on tax credits tied to data centers and recent hacking incidents. If that happens, AI firms could face tighter congressional oversight and more hearings with CEOs.
Anthropic pledged to put 5 percent of its shares into a Public Benefit Fund and lobbied for a federal AI safety bill with strict trigger thresholds. The draft bill would only allow the government to intervene if damage exceeded $1 billion or if more than 50 people were killed or seriously injured. The move changes both its planned public-good share allocation and its proposed safety approach by making regulation conditional on very high-severity harms rather than earlier risk signals.
Anthropic said it will set aside 5% of its shares to fund donations for public-benefit purposes, excluding any use for lobbying or political donations. The commitment is 5% of its shares, with an additional 2% potentially allocated at the board’s discretion. This adds a formal philanthropic funding mechanism as Anthropic pursues a potential IPO that could value it above $1tn.
Every AI story that matters,
in your inbox by 8am.
TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the
day in two minutes. Follow companies and topics for alerts, or get the
briefing in Slack. Free, no spam, unsubscribe anytime.
Reading TLDRocket needs no cookies, and the readership counts we rely on come from
our own cookieless analytics. Google Analytics is the exception: it sets cookies and
reports to Google, so it stays switched off until you allow it. You can change your
mind any time from “Cookie settings” in the footer.
Strictly necessary
Session security and form protection (tldrocket-session,
XSRF-TOKEN, 2 hours). The site cannot work without them,
so they need no consent.
Always on
Google Analytics 4 (_ga,
_ga_<id>, up to 2 years). Measures which
stories and sections readers use. Google acts as a third-party processor and may
store the data outside the EU. No advertising, no profiling, no data sold.
