AI News for October 23, 2026
Browse any day on a calendar, install it to your home screen, read offline.Read in the app →-
1TechCrunch06:08
OpenAI working on new $30B raise at $1.25T valuation ahead of IPO
OpenAI is working on a roughly $30B round that would value the company at $1.25T, which would make it the most valuable private company on record. The raise is being lined up ahead of a planned public listing, signaling how much capital the company expects to need for compute and infrastructure commitments.
Read original ↗ -
2TechCrunch21:11
DeepSeek reportedly raises $15B days after US restrictions
DeepSeek has reportedly closed a $15B round, among the largest ever raised by a Chinese startup. The funding comes days after the US ordered federal agencies to remove the company's technology on national security grounds, underscoring that domestic capital is backing the lab despite the restrictions.
Read original ↗ -
3TechCrunch03:00
OpenAI makes its Jev competitor widely available as decision model competition heats up
OpenAI has opened up general access to its decision model, its answer to Jev. The move turns what had been a limited release into a broadly available product and puts the two systems in direct competition in a category that is drawing growing attention.
Read original ↗ -
4TechCrunch21:53
Claude Duet app is Anthropic's bet on agents
Anthropic has introduced Claude Duet, an app built around agentic workflows rather than single chat exchanges. It is being rolled out through a waitlist for Pro and Max subscribers, making it a paid-tier test of how far users will delegate work to agents.
Read original ↗ -
5TechCrunch02:43
Microsoft’s Mustafa Suleyman says its AI outperforms doctors on hard cases as he chases ‘medical superintelligence’
A study in Science reports that Microsoft's AI diagnostic orchestrator outperformed generalist doctors on difficult cases, with the company claiming accuracy up to four times higher. TechCrunch notes that most of the gain appears to come from the underlying models rather than Microsoft's orchestration layer. The Verge also published an interview with Microsoft AI CEO Mustafa Suleyman on the same results.
Read original ↗ -
6The Verge05:51
Canva’s first AI image model plans designs before generating them
Canva has launched Canva Craft, its first in-house image generation model, which uses a two-stage architecture that plans a design before rendering it. The company says the approach gives more precision, consistency and control, and is pitching it against image models from OpenAI, Microsoft and Google. TechCrunch reports it is rolling out to all users with lower costs than Canva's previous third-party options.
Read original ↗ -
7The Verge23:50
OpenAI, Anthropic, and Nvidia had record lobbying quarters
Federal disclosures show OpenAI, Anthropic and Nvidia each posted their highest-spending lobbying quarter between July and September. Nvidia led at $2.2M, followed by Anthropic at $2.1M and OpenAI at $1.5M, though the totals remain below those of longer-established Big Tech firms in Washington.
Read original ↗ -
8The Verge04:32
Meta's AI could still scrape news sites, legal experts say
Meta has revised its business terms to drop language that appeared to let it train AI models freely on news sites. Legal experts told The Verge the edit does not go far enough, because broad wording about content "associated with your products or services" would still permit scraping.
Read original ↗ -
9TechCrunch00:47
Chinese hackers spied on thousands of companies after breaching IT giant ConnectWise
Microsoft attributes a breach of IT management vendor ConnectWise to a China-aligned group that then spied on thousands of downstream customers. According to Microsoft, the attackers used AI agents to move through victim environments at unusual speed, an early large-scale case of agentic tooling in an intrusion campaign.
Read original ↗ -
10TechCrunch00:00
Uber puts all 30,000 employees on ChatGPT Work
Uber is rolling out ChatGPT Work to its entire workforce of about 30,000 after running a much smaller pilot. It is one of the larger single deployments of OpenAI's enterprise agent product to date and a signal of how quickly pilots are converting to company-wide contracts.
Read original ↗ -
11TechCrunch03:07
1Password launches a way for AI agents to log in on your behalf without seeing your password
1Password has released a feature that lets AI agents sign in to a user's personal accounts without exposing the underlying credentials. It addresses one of the main blockers for agents acting on the open web, where the usual workaround has been handing over passwords outright.
Read original ↗ -
12The Verge03:45
Sure, let’s smoosh SpaceX and Tesla together, who cares
SpaceX and Tesla have set up special committees to discuss a merger, at a point when SpaceX is preparing to borrow another $40B to spend on Nvidia chips. The Verge argues the combination mainly serves to spread the credit risk of that AI compute buildout across Musk's companies.
Read original ↗ -
13TechCrunch08:45
Left alone with nothing to do, AI agents started playing
MIT researchers placed 13 AI agents on a virtual island with no assigned tasks and watched them invent games, hold a ritual over a lost ring, and form social routines. The agents later performed better on actual work, which the researchers read as evidence that unstructured interaction can improve agent behavior.
Read original ↗ -
14The Verge02:15
In Meta’s version of the future, I’ll have my Muse call your Muse
Mark Zuckerberg is positioning Meta around an autonomous network of AI agents that shop, work and negotiate with other people's agents. The Verge questions whether users want to communicate primarily through bot proxies, and notes how central the vision has become to Meta's pitch to investors.
Read original ↗ -
15The Verge04:24
Meta turns Muse sharing back on
Meta has restored sharing in its Muse assistant after suspending it when researchers at Claroc showed that public conversation links were easy to discover. The exposed pages included a marketing firm's internal revenue review and a small business owner's loan application, a reminder of how much sensitive data users put into assistants.
Read original ↗ -
16TechCrunch23:03
Spotify's current AI bet is about being the music service that agents choose
Spotify co-CEO Gustav Söderström says the company's AI strategy is to make itself the music service that third-party agents call when a user asks for music. The framing treats agents as a new distribution channel rather than a competitive threat to streaming platforms.
Read original ↗ -
17TechCrunch21:26
Nirva raises $33M for a pendant that only listens to you
Nirva has raised $33M for an ambient AI pendant designed to capture only the wearer's speech, a response to the privacy objections that have dogged always-on recording devices. The $249 device is due to ship by the end of the year.
Read original ↗ -
18The Verge01:21
Someone hacked the Switch 2 by using an old game and AI
A modder known as pepsipu says OpenAI's Codex helped build "ns2jb," a jailbreak for the Nintendo Switch 2 that chains through an old game title. A demo shows a modified Mario Kart World with custom characters, and the case is an early public example of coding models being used for console exploit work.
Read original ↗ -
19The Verge00:16
Wildwood’s director says AI is ‘antithetical to everything that we are at this studio’
Travis Knight, director of Laika's stop-motion feature Wildwood, tells The Verge that generative AI runs counter to the studio's craft-based approach. The film took five years and more than 230 handmade puppets, and the comments add to a continuing split in animation over where AI tools belong.
Read original ↗
