Author: Tor Fosheim

OpenAI aims for another model in January, in addition to one this week

OpenAI will stay in Code Red for the better part of a month -- until a new model arrives in January
We can expect not one, but two models from OpenAI in the near future. (Picture: generated)
According to The Wall Street Journal, Code Red at OpenAI will extend well into next month.

The force behind it was the release of competent models from Google and Anthropic, making it look like OpenAI was losing the edge in the AI race.

Continue reading “OpenAI aims for another model in January, in addition to one this week”

Google demos smart glasses with Android XR, set to debut in 2026

The glasses without a screen will arrive first, as of "next year."
Google’s in-specs screen is impressing reviewers, but it’s the screenless glasses getting released first. (Picture: Google)
Google has officially taken the lid off «Project Aura,» inviting a whole host of websites to demo it — and doing their own bit in The Android Show, XR Edition on Youtube.

They are mostly concerned with the glasses with internal screens, that can run bog standard Android apps as well as Android XR apps — and provides you with information right inside the glasses.

These spectacles, while impressive, connect via wire to a puck in your pocket that serves as a battery and trackpad in one, and use a phone or laptop for computing power, but they don’t have a release date as of yet.

Continue reading “Google demos smart glasses with Android XR, set to debut in 2026”

Nvidia gets all clear from Trump Commerce to sell H200 chip in China

Nvidia gets approval to ship powerful H200 chips to China.
Nvidia’s more powerful H200 chips can now be sold in China. (Picture: Nvidia)
The powerful chip can be sold to vetted partners with a tax of 25% to the U.S. government, Trump said in a statement.

Nvidia’s H200 is a much more advanced chip than the custom H20 that the company was allowed to export to China earlier, with the newer Blackwell chips being only about 1.5 times faster, Reuters notes.

The H20 chips were recently banned in China, where the authorities instead opted for Huawei chips for data center supplies.

Continue reading “Nvidia gets all clear from Trump Commerce to sell H200 chip in China”

NYT sues Perplexity for copying content, after cease and desist order

Copyrights might well trump AIs retrieval practices, bets the NYT.
The Times says Perplexity is copying their journalism and delivering it without permission. (Picture: Adobe)
The New York Times sent a cease and desist order to Perplexity in 2024, but the company has persisted with copying NYT content in their responses, the lawsuit alleges.

Perplexity still generates outputs that are «identical or substantially similar to» content from the Times, writes CNBC, and sometimes even hallucinates responses that get attributed to them, writes Reuters.

— While we believe in the ethical and responsible use and development of AI, we firmly object to Perplexity’s unlicensed use of our content, says NYT spokesperson Graham James.

Perplexity seems unfazed by the lawsuit, saying in a statement that:

— Publishers have been suing new tech companies for a hundred years, starting with radio, TV, the internet, social media and now AI. Fortunately it’s never worked, or we’d all be talking about this by telegraph.

The NYT has previously also previously sued OpenAI for infringement.

Read more: The actual complaint, NYT announcement, writeups on Reuters and CNBC

Head of ChatGPT says not doing ad tests, despite a Target notice this week

These were not ads, OpenAI says, and anyway, they have been shut off.
Many users were confronted with this graphic at the end of ChatGPT’s responses this week. OpenAI says they are now stopped, and that they weren’t ads. (Picture: screenshot)
People have been complaining about ads being sneakily inserted in their ChatGPT responses late this week, with a «Shop for home…» Target link showing up in their responses.

This comes hot on the heels of an engineer discovering ad code in the Android app — which left a lot of people cautious of such inserts.

It happened to enough people to notice, along with this writer, but it has since been turned off, says Mark Chen, Chief Research Officer at OpenAI:

Continue reading “Head of ChatGPT says not doing ad tests, despite a Target notice this week”

OpenAI loses privacy fight for ChatGPT message logs in NYT lawsuit

OpenAI is on the verge of losing its fight to keep their users' chat logs private, instead of turning them over to the NYT.
Your logs belongs belong to us, the NYT lawyers say, and the court agrees. (Picture: Adobe)
OpenAI has been fighting tooth and nail to preserve the privacy of their users’ messages in the lawsuit brought by The New York Times in 2023.

A new judgement could now mean they have to turn over more than 20 million chat logs, and many more messages, from the chatbot, reports Reuters.

The logs themselves should be anonymized by OpenAI in a way that pleases the court, but their content could be easy to pin down, and OpenAI has promised to appeal to the presiding Judge.

This is merely the discovery phase of the ongoing trial, where lawyers for the NYT have said the messages are necessary to discover whether ChatGPT did indeed copy verbatim text from them.

OpenAI must now first anonymize the logs, and then submit them to the court, and NYT’s attorneys, seven days later.

Read more: Reuters has the scoop.

Google partners with Replit to bring «vibe coding» to the enterprise

Replit tightens its integration with Google models and Cloud.
Vibe Coding is comping for the enterprise. (Picture: Google, modified)
Decade old Replit has a valuation of $3 billion dollars and is a «leader» in the AI Vibe coding space, writes CNBC, and they are now tightening their integration with Google Cloud and the Gemini models.

—The goal for us, and Google, is to make enterprise vibe-coding a thing, Replit founder and CEO Amjad Masad said; — We want to show the world that these tools are actually going to transform businesses and how people work.

Under the new agreement, Replit will expand its Google Cloud use and «further integrate Google’s models into its platform,» Google writes on the deal.

Replit will gain access to all of the Gemini models, and the deal will «help enterprise customers embrace vibe coding.»

— Our mission is to enable the next billion software creators — from hobbyists to entrepreneurs to enterprises, Masad said.

Read more: Google’s announcement, writeup on CNBC.

Google’s Gemini 3 Deep Think debuts for Ultra users

Deep Think incorporates the code that won the math olympiad, and is reserved for heavy lifting.
Gemini 3 Deep Think brings some serious AI muscles to tackle the toughest problems. (Picture: Google)
Google just launched it’s most capable model to it highest paid tier, ready to take on the most confounding probblems.

It incorporates the solutions that won gold at the International Mathematical Olympiad and beat the ICPC coding contest, and carries on its duties doing parallel reasoning, letting it try several approaches to a problem at once.

Continue reading “Google’s Gemini 3 Deep Think debuts for Ultra users”

Apple’s interface chief leaves for new Meta AI design studio

Alan Dye, the Liquid Glass chief is leaving Apple for greener pastures at Meta.
Alan Dye introducing the Liquid Glass design he oversaw at Apple, in June, 2025. (Picture: Apple)
Alan Dye, who has been heading up user interface design for Apple for the last ten years, is officially leaving to lead AI design for Meta, reports Bloomberg.

He brings with him his deputy at Apple, Billy Sorrentino, and will oversee a new studio that encompasses hardware, software and AI integration across all of Meta’s products — including future AI devices.

— Our idea is to treat intelligence as a new design material and imagine what becomes possible when it is abundant, capable, and human-centered, writes Mark Zuckerberg on Threads, not being short of ambition.

He goes on to say that Meta will «elevate design,» and put together people with talents in «craft, creative vision, systems thinking, and deep experience building iconic products that bridge hardware and software»

The potential is «enormous,» Zuck says, for AI glasses and «other devices» to change how we connect with «technology and each other.»

Read more: Scoop by Bloomberg, writeups by The Verge and 9to5Mac.

Micron is pivoting to AI chips, will kill off Crucial consumer brand in 2026

Micron is pitching towards the higher margin AI memory business, and is killing off the beloved Crucial consumer brand.
AI chips are much more profitable than consumer grade products, so Micron is changing turf. (Picture: Micron)
Micron has decided that selling consumer memory chips isn’t profitable enough, in an AI world screaming for as many HBM-chips as at all possible.

— The AI-driven growth in the data center has led to a surge in demand for memory and storage, says Sumit Sadana, EVP and Chief Business Officer at Micron Technology in a press release.

Continue reading “Micron is pivoting to AI chips, will kill off Crucial consumer brand in 2026”

Google linking AI Overviews with AI Mode, further worrying web publishers

Transitioning seamlessly from Overviews to AI Mode instead of websites, Google is keeping more users on its own pages.
Google will send you to AI Mode instead of to websites. (Picture: Screenshot)
Rolling out globally on mobile as of Monday, Google is giving its AI Overviews a little extra depth.

The idea is that sometimes the user is satisfied with a quick overview as an answer to a query, but sometimes it brings up more questions and requires a little more digging.

Therefore, the AI Overview now sometimes comes with an input field for AI Mode at the bottom of the screen, which will be able to give more comprehensive answers – rather than sending you to a website.

— It’s one seamless experience: a quick snapshot when you need it, and deeper conversation when you want it, says Vice President of Product for Google Search, Robby Stein, on x.com:

Continue reading “Google linking AI Overviews with AI Mode, further worrying web publishers”

Anthropic begins testing IPO waters, hires specialist firm that helped Google

Anthropic has started laying the groundworks for an initial public offering.
With a valuation north of $300 billion, Anthropic would be one of the world’s largest IPOs. (Picture: generated)
Just as the AI lab is chasing a funding round said to value it at more than $300 billion, The Financial Times reports they are also considering going public.

This stems from hiring Wilson Sonsini, which assisted Google and LinkedIn in their initial public offerings, and apparently having talks with «big investment banks.» They have also hired Krishna Rao, the former Airbnb executive who played a key role in their IPO in 2020, CNBC says.

The timeline does not seem clear, while the report indicates that it could happen as soon as 2026, it also notes that the talks are in informal and early stages.

Anthropic has 300,000 business and enterprise customers, and is aiming to triple its annual revenue to around $26 billion next year, Reuters writes.

Its IPO could be one of the biggest ever, and would be a test of investors’ appetite for loss-making companies with huge investment bills, FT notes.

Read more: Scoop by The Financial Times, writeups from Reuters and CNBC.

SoftBank sold $5.83 billion of its Nvidia stock in October: Here’s why

People were thinking SoftBank had seen a bubble forming. Far from it, according to its CEO.
The massive stock sale fueled a lot of rumors and worries. (Picture: generated)
Known for its massive stakes in emerging technologies, SoftBank’s disclosure of the sale caused a 2% dip in Nvidia’s share price and inflated talk of an AI bubble.

Now CEO Masayoshi Son has revealed that it certainly wasn’t because they were taking profit before any downturn. On the contrary:

— I was crying to sell Nvidia shares, he told the FII Priority Asia forum in Tokyo, and says he «didn’t want to sell a single share,» adding that «I respect Jensen, I respect Nvidia so much,» according to Business Insider.

Continue reading “SoftBank sold $5.83 billion of its Nvidia stock in October: Here’s why”

Altman declares «Code Red» at OpenAI, plans Gemini 3-beating model

OpenAI has put ads on hold until it can release a Gemini 3-beating model.
All other initiatives are on hold as OpenAI prepares its next model. (Picture: generated)
Competition is heating up in the AI chatbot market, as highlighted in the last weeks, with new, capable models from Google, Anthropic, and a new Codex Max from OpenAI.

This has now caused Altman to delay other initiatives, such as ads, to focus on making a better ChatGPT, paywalled The Information writes, citing an internal memo.

They are apparently planning to release a new reasoning model next week that will be «ahead of Gemini 3.» But this needs a little more polishing on the «experience.»

Just last week, a developer revealed ad code in the latest ChatGPT beta — meaning that their work on ads was fairly advanced and almost ready to ship.

This work is now on the back burner, at least until next week, when OpenAI hopes to reclaim their crown.

Read more: The Information (paywalled), and Reuters. Discussion on r/Singularity.