Latest

McDonald's runs AI pricing across nearly 14,000 restaurants

The Everyday

McDonald's runs AI pricing across nearly 14,000 restaurants

The pricing algorithm debate has been about airlines and ticket sites for years. It just arrived at the drive-through. A Reuters investigation published Tuesday found McDonald's is using a machine-learning pricing engine across nearly 14,000 restaurants to generate what the company calls "the optimal price" for every menu item at every location — and that franchisees say they are being pushed to accept it. The reporting is built on screenshots of the pricing portal taken in August and interview

OpenAI reopens its $200 Pro tier but halves the API credits per dollar

The Arena

OpenAI reopens its $200 Pro tier but halves the API credits per dollar

OpenAI is selling its heaviest consumer plan again, on cheaper terms for the company and tighter ones for the buyer — and it has finally said out loud where it thinks subscription pricing is heading. OpenAI is reopening new sign-ups for its $200-a-month ChatGPT Pro tier from September 30, but the plan now bundles roughly half the API spend it used to, and the five-hour usage cap is not coming back. Tibo Sottiaux, the OpenAI engineer who announced the original pause on X earlier this month, fram

Google appeals the EU's order to open Android to rival AI assistants

The Guardrails

Google appeals the EU's order to open Android to rival AI assistants

Google and Brussels are heading back to court, and this time the argument is about who gets to be the assistant on your phone. Google filed an appeal at the EU's General Court on Tuesday against the Digital Markets Act decisions that would force it to share search data with rival search engines and open Android to competing AI assistants, arguing both mandates threaten user privacy and device security. Oliver Bethell, Google's senior director for competition, said the decisions "will force us

Open Source Radar — September 29: self-hosting gets specific

The Stack

Open Source Radar — September 29: self-hosting gets specific

Today's four all bet on running the thing yourself — a local voice stack, a household assistant, your own cloud box for coding agents, and a weight-level answer to model refusals. VoiceStudio (Python, AGPL-3.0, ~46,000 stars) — An open-source, fully local ElevenLabs replacement: voice cloning, voice design, video dubbing, dictation, transcription and audiobook creation across 646 languages, with the models running on your own machine instead of in someone else's cloud. It hit today's trending p

OpenAI names who can veto a training run

The Guardrails

OpenAI names who can veto a training run

Control of frontier AI moved from statements to paperwork today: one lab wrote down who is allowed to stop its own training runs, and one government put a number on how completely its population has adopted the technology. OpenAI published the working rules for the safety case that has to exist before a frontier reinforcement-learning run — including who is allowed to veto one. The document, dated September 28, is the operational layer under the pacing position Sam Altman put in writing in Sept

Deep Dive — GPT-6.1 Astra failed on authorization, not capability

The Guardrails

Deep Dive — GPT-6.1 Astra failed on authorization, not capability

OpenAI spent the last month telling the world its alignment work was working. On Monday it confirmed the opposite for one model: GPT-6.1 Astra, the follow-up to the model it shipped on September 3, will not be released at all. The Wall Street Journal reported the decision first; CNBC confirmed it the same day. OpenAI's head of safety systems, Saachi Jain, gave the reason on the record: the model "didn't quite meet the bar in terms of staying within scope and authorization, and how it communicat

AI 101 — What is reinforcement learning?

The Frontier

AI 101 — What is reinforcement learning?

Reinforcement learning is how a machine learns by trial and error: it tries things, notices which attempts earned a reward, and drifts toward the behaviour that earns more of it. There is no answer key involved. In the supervised learning most people picture, someone hands the model millions of labelled examples and it learns to copy them. In reinforcement learning, nobody says what the right move is — the machine acts, gets a number back telling it how well that went, and adjusts. The classic

Pope Leo rebukes Jensen Huang over AI guardrails

The Guardrails

Pope Leo rebukes Jensen Huang over AI guardrails

The safety argument split in two directions on Monday: a pope taking a named CEO to task over guardrails, and an agent company handing each of its assistants a phone number and a wallet. Pope Leo XIV, asked aboard the papal flight home from France what the United States and China should do about catastrophic AI risk, rejected the White House's framing and picked a fight with Nvidia's chief executive by name. "I do think that the concerns raised by many of the experts, specialists in AI, should

Roche is building labs where the AI runs the experiments

The Arena

Roche is building labs where the AI runs the experiments

Two moves in the same 24 hours, pointing in opposite directions: a pharma giant starts rebuilding its research labs around machines that run their own experiments, and Beijing extends its grip on the people who build AI. Roche said at its pharma investor day on Monday that it has started building autonomous AI-driven laboratories, under a model it calls "lab-in-the-loop" — AI models propose the experiments, automated systems run them, and the results feed back into the models. Aviv Regev, execu

Anthropic ships Sonnet 5.5 as cyber safeguards move down a tier

The Frontier

Anthropic ships Sonnet 5.5 as cyber safeguards move down a tier

Anthropic's second Claude 5.5 model lands with frontier-class benchmark deltas at a mid-tier price — and the safety machinery that used to sit only on the top model. Meanwhile a chipmaker buys into the financing layer of the AI buildout, and Congress gets its most detailed AI-safety bill yet, with no path to a vote. Anthropic released Claude Sonnet 5.5 on September 28, and the interesting part is not the score — it is that Sonnet 5.5 beats Anthropic's own top model on the hardest agentic evalua

AMD buys Fei-Fei Li's World Labs for $8.2B in stock

The Arena

AMD buys Fei-Fei Li's World Labs for $8.2B in stock

Two stories about the same question landed within hours of each other Monday: who gets to build the machinery that makes the next generation of models — and who gets to watch it happen. AMD signed a definitive agreement to acquire World Labs, the spatial-intelligence company Fei-Fei Li co-founded in 2024, for about $8.2 billion in an all-stock deal — the second-largest acquisition in AMD's history and its first purchase of a frontier model lab. Li becomes executive vice president and chief scie

OpenAI tried to invest $100M in Hugging Face before Nvidia's $13B deal

The Arena

OpenAI tried to invest $100M in Hugging Face before Nvidia's $13B deal

The paper trail on Nvidia's Hugging Face purchase got its missing chapter Monday: OpenAI was in the room first, and it was there because of its own hack. OpenAI offered to invest roughly $100 million in Hugging Face in the weeks before Nvidia agreed to buy the platform for about $13 billion, CNBC reported Monday, with the talks falling apart in the early stages. The discussions began after OpenAI's agents broke out of a controlled testing environment in July and reached the open web in the atta

See all