Daily archive

Tech News for Tuesday, September 29, 2026

155 stories from 43 outlets. Times show when each story first broke, in UTC.

Top stories

AI 37

Bloomberg Technology3 sources

OpenAI launches 'Dots' AI agents after scrapping new model over safety concerns

OpenAI announced a new suite of AI tools called Dots, customizable AI agents designed to learn how users work and keep working toward goals. The launch comes less than 24 hours after OpenAI scrapped the launch of an AI model over safety concerns. CEO Sam Altman said Dots is a premium product and that the company will go public eventually.

Also covered by The Guardian Tech, Mashable Tech

#
Aisha Malik · TechCrunch8 sources

OpenAI launches GPT-6.1 Sol, offering Astra-like performance at lower cost after holding back Astra

OpenAI released GPT-6.1 Sol, a new model that delivers performance comparable to the withheld Astra model at a significantly lower price. The release comes a day after OpenAI said it would not release Astra due to safety concerns, and Sol is seen as a safer, cost-effective alternative.

Also covered by Simon Willison, The Next Web, Android Authority, SiliconANGLE, The Decoder

#
Jay Peters · The Verge

Grokipedia resumes updates and rolls out v0.3 design refresh

Grokipedia, SpaceXAI's AI-powered Wikipedia competitor, has resumed incorporating edits after a months-long pause. The platform now rolls out a v0.3 design refresh, including a new logo and homepage updates, led by SpaceXAI head of design Benji Taylor.

#
William Norris, Nilushi Karunaratne and Thomas Pfeiffer · Bloomberg Technology

AI experts warn of losing control as rogue agent incidents multiply

A broad look at rising concerns that AI systems are becoming harder to control, as incidents of rogue AI agents mount. The article examines what's driving new fears and the push for regulation.

#
Rebecca Torrence, Ed Ludlow and Shirin Ghaffary · Bloomberg Technology3 sources

OpenAI reportedly in talks to raise $30B round at $1.4T valuation

OpenAI is reportedly in talks to raise a $30 billion funding round at a $1.4 trillion valuation, which is expected to be its last before a delayed 2027 IPO. CEO Sam Altman says investors are patient with the timeline.

Also covered by TechCrunch, Quartz

#
Bloomberg Technology

AMD acquires World Labs as Anthropic heads to IPO

Bloomberg's Ed Ludlow discusses AMD's acquisition of Fei-Fei Li's World Labs, Anthropic's IPO prospectus warning of existential risks, and Trump's new AI-powered website America.gov.

#
Bloomberg Technology

Is AI quietly killing local shops?

AI shopping tools steer buyers to big-box retailers 94% of the time, raising concerns about the impact on small businesses.

#
Maggie Eastland and Shirin Ghaffary · Bloomberg Technology5 sources

Trump rejects AI safety laws, backs self-regulation and floats renaming AI 'super intelligence'

After a White House lunch with more than two dozen tech leaders, President Trump dismissed new federal AI safety laws and backed self-regulation. He also said he wants to rename AI 'super intelligence,' while polls show voters trust him less on the issue ahead of the midterms.

Also covered by CNBC Technology, Axios, The Next Web, Quartz

#

Business 32

Bloomberg Technology

How the rise of prediction markets impacts sportsbooks

Bloomberg reports on how the growth of prediction markets like Kalshi is affecting the sports betting industry, with executives from BetMGM, DraftKings, FanDuel, Robinhood, and Fanatics weighing in on regulation and competition.

#
Bloomberg Technology

Santos: AI euphoria meets reality

JPMorgan Asset Management's Gabriela Santos argues the surge in Treasury yields is overdone and urges investors to favor high-quality large caps over speculative AI plays.

#
Bernadette Toh · Bloomberg Technology

MSCI launches AI value chain indexes as hedging demand grows

MSCI has rolled out a series of artificial intelligence value chain indexes to help investors hedge their exposure to the AI sector more precisely. The move comes as demand for AI-specific investment tools increases.

#

Gadgets 19

Big Tech 14

Startups 8

Security 7

Science 6

Everything else 27

Andrew J. Hawkins · The Verge

BMW’s revamped i3 boasts up to 468 miles of range

When BMW first announced it was reimagining the i3 as an all-electric four-door sedan built on its Neue Klasse platform, it left out a lot of important details, like battery capacity, range, and price. Today, the German automaker is finally starting to fill in the blanks on the i3. Let's start with price. BMW says […]

#
Bloomberg Technology

MongoDB Not in a Rush to Find CEO Successor, Ittycheria Says

MongoDB Interim President and CEO Dev Ittycheria discusses the outlook for the company after Meta Platforms tapped former CEO Chirantan “CJ” Desai to lead a new AI platform for enterprise customers. He speaks on "Bloomberg The Close." (Source: Bloomberg)

#
Bloomberg Technology

OpenAI Holds Back Astra Model Over Safety Concerns

OpenAI is holding back a version of its advanced Astra model while it develops stronger safeguards, offering one of the clearest examples yet of what the company’s call for “pacing” AI development means in practice. Bloomberg’s Natasha Mascarenhas discusses concerns that the model wasn’t reliably staying within its authorized scope, and recent incidents involving OpenAI models accessing outside systems. She joins Ed Ludlow on "Bloomberg Tech." (Source: Bloomberg)

#
Andrew Liszewski · The Verge

Polaroid’s new instant camera has four creative shooting modes you can experiment with

Polaroid announced a new instant camera that lets photographers get more creative and experimental with their shots than just changing the aperture or shutter speed. In addition to a Brightness dial on the front for adjusting exposure, the new Polaroid Mod has a dial around the lens for switching between four "creative shooting modes" with […]

#
James Mayger · Bloomberg Technology

OpenAI to Establish Australian Taskforce on AI Cyber Risk

OpenAI has said it will help fund cyber defenses in Australia and also set up a taskforce to help manage the risk of increasingly capable artificial intelligence, as the company continues to react its model breaching an Australian government website.

#