New: OpenAI quietly rolled back ChatGPT's model router for free users last week. The router—which was costly for the company to serve, but aimed to improve answer quality—was thought to be hurting engagement, one person familiar with the matter told WIRED.

OpenAI is losing its chief comms officer; Hannah Wong had been there since 2021. scoop from @mzeff.bsky.social

Loved this piece about Taiwan’s public transit keychains from @lmatsakis.bsky.social, and especially this one collector: “Later, I had a lunchbox-shaped one, but looking at it too long would make me hungry, so I never really used it. I just kept it as a toy.” www.wired.com/story/icash-...

NEW: Four sources tell @mzeff.bsky.social that OpenAI has become reluctant to publish research on the negative economic impacts of AI, including job displacement. At least two members of OpenAI's economic research team have recently quit over that perceived pullback.

An OpenAI staffer has quit, claiming that the company has backed off from publishing negative economic research about AI. scoop from @mzeff.bsky.social www.wired.com/story/openai...

SCOOP: As OpenAI faces increased scrutiny over how ChatGPT responds to distressed users, a research leader spearheading that work announced her departure, WIRED has learned. Andrea Vallone, the head of model policy, will leave the company by the end of 2025.

Scoop w/ @peard33.bsky.social and @mzeff.bsky.social: OpenAI locked down its San Francisco offices today after receiving a threat from an activist previously associated with Stop AI: www.wired.com/story/openai...

A $100 million pro-AI super PAC has picked its first target. @mzeff.bsky.social talked to New York congressional candidate Alex Bores about what it's like to be in Andreessen Horowitz's crosshairs: www.wired.com/story/alex-b...

NEW: Trump Takes Aim at State AI Laws in Draft Executive Order All the deets + more w/ @mzeff.bsky.social (it's his first WIRED byline yay!!!!) www.wired.com/story/trump-...

Excited to announce I’m joining @wired.com as a senior writer covering the business of AI! I’ve long been a fan of WIRED, but its recent coverage has felt especially impactful. I'm thrilled to work with this talented team, and continue reporting on AI as it reshapes Silicon Valley and the world.

Personal news: today is my last day at TechCrunch. Two years ago I moved to SF to be closer to the people building AI. I'm so happy I did — it's been a wild, fascinating, and exciting ride. Excited to keep reporting on AI and the industry's leaders in my next role. More soon!

Ex-OpenAI safety researcher Steven Adler got his hands on one of ChatGPT's famous cases of delusional spiraling. Today he published his analysis of Allan Brooks' incident in May, first covered by @nytimes.com “I’m really concerned by how OpenAI handled support here,” Adler told me.

You know how every podcast is a video now? Well ... If you like seeing my face when I talk about tech news, then today is your LUCKY DAY. (Better still, you get to see @kirstenkorosec.bsky.social + @mzeff.bsky.social) techcrunch.com/2025/09/19/t...

Scoop: OpenAI reorganized its Model Behavior team to be closer to core model development, TechCrunch has learned. Plus, the founding leader of Model Behavior, Joanne Jang, is starting a new unit called "OAI Labs," focused on "novel interactions" between humans and AI.

SCOOP: A Scale executive Alexandr Wang brought over to help run MSL has departed Meta after just two months. Plus, researchers in TBD Labs have complained that Scale AI's data is low quality, and are using its competitors, Surge and Mercor, to train next-gen AI models.

NEW: Meta is updating how it trains AI chatbots such that they can't have inappropriate romantic conversations with teenage users. This follows a bombshell report that found Meta was allowing its AI chatbots to have sensual chats with teens. Meta calls these "interim" changes.

Elon Musk said Grok 4 is “maximally truth-seeking.” But if you ask Grok 4 questions about controversial topics, one of its first steps is to search for what Musk has said publicly about the topic, and then it shapes its response around that. Yes, really: techcrunch.com/2025/07/10/g...

NEW: Cloudflare has a big idea to give publishers a lifeline in the AI era. It's called Pay per Crawl, and it forces AI companies to pay publishers every time they scrape a website. New websites set up with Cloudflare will also block all AI crawlers by default now.

techcrunch.com/2025/06/27/c... Worked with @mzeff.bsky.social on this overview of the AI moratorium, the provision in Trump's "Big Beautiful Bill" that would block states from regulating AI for a decade. Read arguments for and against, the laws at stake, and what it all means. @techcrunch.com

“The window to put in place guardrails is rapidly shrinking given how fast this technology is evolving,” Sen. Andrew Gounardes, told @mzeff.bsky.social. In order to prevent disaster scenarios, the bill requires security reports, the reporting of safety incidents and penalties for non-compliance.

NEW: Anthropic CEO Dario Amodei claims that AI models today hallucinate less than humans do. I asked Amodei whether hallucinations were a limitation to AGI at Anthropic's Code with Claude event. He argued it's not, and claimed to see no "hard blocks" on what AI can achieve.

NEW: At I/O 2025, Google laid out a future where AI agent bring users the web in whatever medium they like. This pitch of AI-enabled Search is much more convincing than last year's AI Overviews launch, however, it likely has larger implications for the web.

As people fill in their seats at Google I/O 2025, employees are vibecoding productivity apps onstage. Help

This is your weekly reminder to not use generative AI in court techcrunch.com/2025/05/15/a...

NEW: Mark Zuckerberg spoke in SF on Tuesday and laid out his plan to test thousands of AI generated ads across Meta's apps. This black box ad tool has obvious implications for marketers, but also users on Meta's platforms, which are quickly filling up with AI slop

Mark Zuckerberg is talking at Stripe’s conference, and was just asked about Apple: “Tim’s had a bad week, I’m not gonna pile on… but I like Sundar”

sf is so warm today im wearing shorts and a t shirt, breaking my unspoken rule of never wearing both simultaneously. all men I know have nonsensical rules like this.

NEW: A study led by Cohere researchers claims LM Arena helped a select group of top AI labs — including Meta, Google, Amazon, and OpenAI — game its popular Chatbot Arena. LM Arena co-founder Ion Stoica tells me the study is full of "inaccuracies" and "questionable analysis."

Meta's first AI developer event, LlamaCon, is official wrapped. It felt more about OpenAI than developers. The company released a Meta AI chatbot app and a Llama API service, both of which seem poised to undercut OpenAI's core businesses.

NEW: OpenAI has been on a quest to acquire a vibecoding startup. Why? Just look at the rapidly rising revenues of Cursor and Windsurf, which we unveil in here. But buying one is easier said than done.

NEW: OpenAI's reasoning models are getting a lot better, but their hallucinations are getting worse. I talked to researchers at Transluce AI who identified some odd hallucination problems with the o3 model, which said it could "run code on a 2021 Macbook Pro." techcrunch.com/2025/04/18/o...

NEW: OpenAI says it's not releasing the standard safety report — AKA a system card — for GPT-4.1, that it has released for most of its previous AI models. The company's excuse? OpenAI says GPT-4.1 is not a frontier AI model.

Meta saying it just happened to release a version of Llama 4 that performs well on LMArena seems harmless, until you actually look at the responses this thing gives and realizes no one would ever want an AI chatbot that talks like this.

NEW: There's no shortage of tech CEOs giving the bull case for how LLMs will reach AGI in a few years. But some AI leaders are more skeptical. Today's AI models still struggle with creativity, which some experts point to as a key limitation to achieving true superintelligence

Pivot back to free expression going well so far

SCOOP: We got details from inside the Humane acquisition. Tuesday was a chaotic day for Humane employees, full of impromptu meetings to explain what the acquisition means, big job offers to work at HP for some, and layoffs for others. Read it @techcrunch.com

Not sure why Dario Amodei and Demis Hassabis are sitting on a very small couch in this interview – while the Economist's EIC is on a very large couch by herself – but I like it

Wanna know when a company is not happy about how much its assets were valued for? Typically, putting the number at the bottom of the press release is a good hint!

Elon's bid to takeover OpenAI is as ridiculous as it is brilliant. Most nonprofits would never have to consider a $97B hostile takeover bid from Elon Musk. And yet, OpenAI may end up defending why it didn't accept. I broke it down with @kylelwiggers.bsky.social for @techcrunch.com

NEW: OpenAI's Operator agent requires too much user intervention to be considered truly "autonomous" today, but it does show where the puck is going. Plus, execs from Instacart and eBay tell me they're thinking about OpenAI's Operator as a new platform.

New: The initial panic around DeepSeek is dying down, but the startup may have a lasting impact on the ways Silicon Valley trains AI models, and how America regulates them.

NEW: I talked to the CPO of Sam Altman's World project, which now wants to link AI agents to your digital identity. As OpenAI releases its first AI agents, Altman's other venture is building tools permitting them to take actions on a human's behalf.

Meta execs obsessed over beating OpenAI’s GPT-4 internally, court filings reveal

NEW: Meta executives and researchers obsessed over beating OpenAI's GPT-4 internally, and dismissed Mistral as a real competitor, according to messages unsealed by a court today. “Our goal needs to be GPT-4,” said an exec. “Mistral is peanuts for us."

New: OpenAI released "Tasks" today, a new way to schedule reminders and recurring requests in ChatGPT. It's a small, but practical step towards agentic AI systems. It's rolling out to paying ChatGPT users this week in beta.

zucks going to get some emotional support high fives from the Trump crowd

NEW: This week, Nvidia unveiled a prototype of its AI avatars for PCs, which have the same screen viewing abilities as Microsoft Recall. I got a chance to try them out, and they're as cool as they are weird.

New for @TechCrunch: The AI wearables war is back baby. Omi, a competitor to Friend that originally had the same name, is launching an $89 consumer device today. It runs GPT-4o and supposedly has a "brain interface." Oh and @kodjima33.bsky.social spent 20% of his funding on a marketing video.

Ok the sphere is sick

Gonna start a thread of stuff from CES 2025 that I’m just calling: “no thank you!” 1. no thank you! techcrunch.com/2025/01/06/g...