OpenAI fired three AI safety researchers tied to the METR/Hugging Face incident and they disputed the reasons while OpenAI reportedly said they mishandled confidential information. Pricing for OpenAI’s GPT-6.1 Sol sets input/output tokens at $12 and $60 per 1M tokens. The updates include broader model rollouts and multiple safety and evaluation releases, plus ongoing debate about agent monitorability and authorization norms.
OpenAI launched a Decisions API to automate routine business classification decisions after TypeSafe’s Jev gained viral traction. OpenAI said its interface can return decisions up to 10 times faster than calling GPT-6 Luna through its general Responses API and priced inputs at $0.10 per million tokens versus Jev’s $0.042. Developers now have two competing “decision” products with different latency and cost tradeoffs, pushing them to adjust how they structure classification prompts and thresholds for human review.
OpenAI fired three AI safety researchers who then published an open letter denying misconduct and disputing the company’s reasons for the dismissals.
OpenAI said the firings followed an investigation and a “pattern of misconduct” involving mishandling research information that went beyond sharing with an outside evaluation group.
The dispute escalates into claims of a chilling effect on employees’ willingness to raise safety concerns and work with third parties, while OpenAI shared internal praise for the researchers and says the decisions were not retaliation.
OpenAI told investors its annualized revenue is lower than an earlier figure shared about it. The updated figure is said to be approaching $50 billion, about $20 billion less than the roughly $70 billion number. This recalculates expectations for how OpenAI funds its large investments and delays plans like an IPO to early 2027.
OpenAI released hundreds of claimed solutions to difficult math problems, but an advisory group says the proofs did not meet field standards focused on human-understandable results and proper formalization. Only 42% of the released proofs had not undergone the suggested formalization process. As a result, mathematicians are questioning reliance on proprietary AI models for trusted proof releases and calling for more formalization support, funding for human mathematicians, and machine-readable links between natural-language explanations and formal artifacts.
OpenAI narrowed its spending gap with Anthropic among companies using OpenRouter model providers. In September, the two rivals split spending almost evenly across roughly 120,000 companies, after Anthropic had accounted for about three-quarters at the start of the year. Companies increasingly choose cheaper models and mix providers, and Anthropic responded with price cuts such as Claude Haiku 5.5.
Brian Chesky said he wishes he were 26 again so he could take advantage of today’s more powerful AI tools to build new ventures. He referenced the year 2008 when he scaled Airbnb and compared it with the post-ChatGPT era. He argues that AI lowers barriers for small teams, enabling tens of millions of new businesses and pushing workers to use the tools rather than treat them only as a threat.
Zvi (Don't Worry About the Vase)·2 weeks ago·
38
● 19 sources
OpenAI released solutions to open math problems, contributing the biggest day in the history of mathematics mentioned in the article. It solved 90 of the top 500 open problems, reaching an average of three hours of Pro-level compute per question. The focus shifts to covering this math breakthrough and to ongoing updates across AI products, adoption, safety policy disputes, and new model releases.
USA Today Co. and its local newspapers sued OpenAI, alleging it copied hundreds of thousands of its articles to train AI models. The filing seeks more than $250 million in damages. The lawsuit adds to a growing set of copyright cases targeting OpenAI over training-data use and could increase legal risk and costs for the company.
Pollo AI added OpenAI-based tools that turn creative ideas into detailed images and cinematic video ads. The integration names GPT-5.6, GPT-6 Astra, and GPT-Image-2.5. This lets creators generate campaign assets directly within Pollo AI, changing the workflow for making ad imagery and video.
OpenAI rolled out GPT-6 to ChatGPT for everyone, adding an “Intelligent UI” interface that can insert interactive graphics, buttons, forms, and small mini-app tools into answers. GPT-6 Instant is reported to start answering 44 percent sooner on average when web search is involved. GPT-6 also switches free users to a smaller “Luna” model and changes how responses appear (with optional visuals that some critics view as more presentation than substance).
OpenAI released 372 major mathematical results, including a proof of Subhash Khot’s Unique Games Conjecture, but the author says few humans yet understand the proofs and are starting a new digestion effort. The work reportedly used about 3 hours of GPT-Pro-level compute on average per solved problem and tested on about 8,000 problems. As a result, mathematicians are rushing to formalize and learn from messy AI-generated proofs while the field shifts toward using AI tooling to help verify and explain progress.
Artificial premiered at the New York Film Festival, with Luca Guadagnino describing its theme of people trying to “play God” as central to the satire of Sam Altman’s rise and ouster at OpenAI. The film ties its narrative to Altman’s power period that included a brief ouster from the AI lab. As a result, the movie keeps close to the factual events it depicts while using that framework to explore power and ambition.
Meta launched Muse and OpenAI launched Dots as always-on AI agents pitched through simple animated mascot interfaces for mainstream users and businesses. Dots are tied to OpenAI subscription tiers of $100 to $200 per month. The focus shifts toward consumer-style agent distribution while raising privacy and security concerns about giving agents access to sensitive data like email and payments.
Researchers allege that OpenAI and Anthropic may have used ChatGPT/Codex/Claude interactions to reproduce or accelerate unpublished work, citing several cases in math and biology. OpenAI published 377 additional math results from an unreleased internal model on GitHub. As a result, the companies face intensified transparency scrutiny, while some researchers plan to avoid proprietary models or switch to open-source alternatives and advocate contract or policy changes to limit training on user data.
Architect Financial Technologies launched Liquid Inference, a router that runs a per-request live auction for LLM inference across providers bidding to serve a named model. The maximum price is locked before the first token is generated. Buyers can swap to an OpenAI/Anthropic-compatible base URL to use marketplace-style pricing with rule-based controls, while providers bid quotes and get payouts through Stripe for metered usage.
U.S. venture capital reached a record $515.8B in deal value in the first nine months, but exits lagged so investors are not getting their capital back as quickly. Venture deal value totaled $515.8 billion with a quarter still remaining, while third-quarter deal value fell about 40% to $98.4 billion. The exit pipeline shifts toward mergers and acquisitions, and without major AI listings or IPOs the report expects weaker liquidity going into 2027.
OpenAI said its GPT-6 models will shift ChatGPT away from text-only answers by adding an “Intelligent UI” that can insert visuals like charts, forms, buttons, and graphics directly into responses. The feature starts rolling out today for ChatGPT Plus, Pro, Business, and Enterprise subscribers, with free users and Go subscribers getting it tomorrow. ChatGPT will use the new interface selectively on more complex requests, while users can still ask for an interactive response even for prompts that would otherwise get plain text.
OpenAI disrupted two AI-enabled influence operations that used false-front journalists and a think tank to spread geopolitical messaging. The operations were identified and disrupted together as 2 separate campaigns. As a result, the AI-assisted messaging efforts were stopped and those channels were disrupted.
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.
