926 summarised stories about AI Policy, each linking back to the original source. Browse all topics →
OpenAI 8 months ago 29 ● 2 sources
OpenAI banned likely China-origin accounts that used AI to research US persons and locations and to plan social-engineering tactics. The activity involved researching US persons and locations and included social-engineering tactics. As a result, those accounts were removed from OpenAI’s services.
OpenAI 8 months ago 36 ● 3 sources
OpenAI banned accounts that used AI to support romance scam workflows, including outreach, translation, victim engagement, and investment-fraud lures. The action involved banning the accounts themselves. This reduces the availability of AI-assisted services for scam operations and signals stricter enforcement against AI-enabled fraud.
OpenAI 8 months ago 40 ● 3 sources
OpenAI banned accounts linked to an influence operation dubbed “Trolling Stone,” which generated comments about an alleged Russian cult leader’s arrest in Argentina using AI. The report says the accounts were linked to a previously unreported operation. As a result, those accounts were removed, and the incident was brought to light as an example of AI-assisted influence activity tied to Russia.
OpenAI 8 months ago 17 ● 3 sources
OpenAI banned accounts linked to the Rybar network that appear to have originated in Russia and used AI to support multilingual influence activity across websites and social platforms. The article says the accounts were linked to the Rybar network. This results in those Rybar-linked accounts being removed from OpenAI services and their AI-enabled influence activity being disrupted.
OpenAI 8 months ago 46 ● 2 sources
OpenAI banned an account linked to a person associated with Chinese law enforcement after detecting AI use tied to influence activity and online harassment operations. The account was banned by OpenAI (date not specified in the article). As a result, access to the account was cut off as OpenAI acted against the planned influence and harassment activity.
The daily briefing
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.
TLDRocket
26 October 2026
Anthropic’s march toward an IPO turned into the defining AI thread of the day, not because the filing was just financial theatre, but because it laid bare the mechanics of how modern model companies intend to survive both scale and scrutiny. In its SEC S-1 for a Nasdaq debut under ticker ANT, Anthropic reported $16.3 billion in first-half revenue and an $80.7 billion net loss for the six months through June 2026, alongside a compute plan that’s hard to ignore: $542 billion to $567 billion on future infrastructure, with “Anthropic Direct” data centers aimed at reducing dependence on major clouds. The strategy is tightly coupled to product, too—Claude isn’t just a chatbot, it’s being pushed deeper into enterprises via first-party apps. More than half of the company’s $100,000-per-year customers use both the Claude platform and a first-party app, and Claude Code’s top 5% of enterprise users spend about 20× more than the average.
Read the full briefing →