The Wire
Daily AI news, learning & projects
News
Timeline
Models
Learning
GitHub
?
Latest
Daily
Social
Videos
‹ Tuesday's edition
·
Daily edition
·
Thursday ›
Wednesday 4 November
All
OpenAI
Anthropic
Google
Other
Policy & law
15:15
AI super PACs spent $250M on midterms with little to show
Pro-AI super PACs spent heavily in the 2026 midterms but mostly backed losing or safe candidates. Leading the Future, backed by $100 million from Silicon Valley billionaires, largely avoided competitive races. Crypto and AI PACs together spent $250 million yet lost most contested races they targeted.
4 sources
TechCrunch
The Verge
Wired
Techmeme
Policy & law
11:01
Texas Democrats' wins cast doubt on booming AI data center buildout
Democrats won the Texas governor's race and a US Senate seat, making data centers a central campaign issue. Governor-elect Gina Hinojosa has pushed for a moratorium on new facilities, and a permit freeze began in September. Texas hosts some of the largest AI data center projects, so restrictions could slow Big Tech's AI infrastructure plans.
3 sources
Wired
MIT Technology Review
Techmeme
OpenAI
12:10
OpenAI acquires developer tools startup Pierre Computer
OpenAI announced it will acquire Pierre Computer Company, a developer tools startup whose technology is already used in Codex, for an undisclosed amount. The deal aims to expand Codex's tools for building, understanding, and reviewing code, with Pierre's Code Storage feature set to be integrated directly into Codex.
3 sources
twitter.com
OpenAI
Techmeme
Google
Developing · since Tue
Google forces AI integration into classrooms, threatening to disable Search access for schools
NEW
Recent reports highlighted significant political and parental backlash against Google's educational mandates, with competitors like DuckDuckGo actively courting frustrated institutions.
Google is tying its core Search functionality to integrated AI features, pressuring educational institutions. Schools that refuse these AI updates face the loss of access to Google Search. This policy shift has drawn criticism from both educators and tech competitors.
4 new · 7 sources
Wired
TechCrunch
The Guardian
Techmeme
Ars Technica
The Verge
The New York Times
Policy & law
16:56
Midterm voters back sweeping restrictions on AI data centers
Voters in the 2026 midterms overwhelmingly backed measures restricting data centers, with 38 of 44 ballot measures passing across ten states. Democrats including Gina Hinojosa won governor races in Texas, Georgia, Arizona, and Ohio, threatening delays to much of planned capacity. Industry groups argue the results reflect an anti-Trump wave, not opposition to AI.
3 sources
TechCrunch
Ars Technica
Wired
AI industry
18:26
Nothing CEO Confirms Launch of AI-First Screen-Free Device
Nothing CEO Carl Pei announced the company plans to release its first AI device in December. The new hardware is designed to leverage artificial intelligence to improve user productivity.
2 sources
TechCrunch
The Verge
Google
13:49
UK says AI will add £55B–£140B by 2030, well below £400B claim
UK AI minister Kanishka Narayan has confirmed the government's own estimate of AI's economic benefit is £55–140 billion by 2030, far below the £400 billion figure ministers cited for two years. That larger number came from a 2023 report commissioned by Google, and researchers partly blame slow government adoption.
2 sources
Techmeme
Wired
OpenAI
00:00
OpenAI launches Images 2.6 with GIF creation across ChatGPT, Codex and API
NEW
The newest reports add OpenAI's claim that Images 2.6 generates images 50% faster than before.
OpenAI has released Images 2.6, bringing animated GIF generation to ChatGPT from prompts or photos, plus faster image generation and new API controls. The update spans ChatGPT, Codex and the API, with the company claiming generation is now 50% faster.
3 sources
Techmeme
The Verge
OpenAI
Meta
18:50
Meta acquires AI coding startup Conductor
Meta has bought Conductor, a startup whose software lets developers run multiple AI coding agents in parallel on the same project. CEO Charlie Holtz said the team will join Meta Superintelligence Labs. The deal adds agent-orchestration talent to Meta's push in AI developer tools.
2 sources
TechCrunch
Techmeme
AI industry
07:30
Qualcomm pushes into AI data centers as smartphone sales slide
Qualcomm's Q4 2026 earnings call outlined plans to grow its AI data center business through partnerships with Amazon and Meta, as smartphone sales fall about 20 percent versus 2025. The shift matters because chipmakers are chasing AI infrastructure demand. Separately, experts warn slow US grid expansion could hinder the AI boom while China builds power capacity faster.
2 sources
The Verge
Techmeme
AI industry
15:58
Uber reports extremely low adoption of AI agents for ride bookings
Uber executives disclosed that only four third-party AI agents booked rides during the third quarter of 2026. This indicates that AI agent integration in ride-hailing remains niche, although usage for Uber Eats orders was significantly higher.
2 sources
TechCrunch
The Verge
Policy & law
Developing · since Tue
Tech stocks jump as Democrats' midterm sweep boosts AI regulation bets
Shares of US tech giants rose after Democrats won the midterms, with investors betting a Democratic Congress will slow AI data center spending and pass frontier AI legislation in 2027. Data center developers face stricter rules after voters backed moratoriums, including bans in at least 14 Ohio municipalities.
2 new · 3 sources
Techmeme
The Verge
The Guardian
OpenAI
08:00
OpenAI pipeline finds and patches 948 Linux kernel flaws
OpenAI detailed an autonomous pipeline that found, validated, and patched 948 local privilege-escalation vulnerabilities in the Linux kernel, using an upcoming cyber model at roughly $1.2M in API costs. It shows security research can run as a systematic process rather than one-off investigations.
2 sources
Techmeme
OpenAI
Policy & law
10:45
56% in NBC exit poll say AI and data centers shaped their vote
An NBC exit poll found 56% of voters said AI and data centers were a crucial or major factor in their Congressional vote choices. Some candidates pledged to halt AI infrastructure development. The result signals significant public discontent with data center expansion.
1 source
Techmeme
Policy & law
06:05
Oracle scraps $7B chip lease deal with Tencent after US pressure
Oracle has abandoned a roughly $7B, five-year deal to lease about 100,000 advanced AI chips to Tencent, according to Financial Times sources. The move followed pressure from the Trump administration, highlighting growing tensions for US AI infrastructure firms doing business with Chinese tech giants.
1 source
Techmeme
Nvidia
05:55
Chinese AI firms' $70B Nvidia chip orders face long delays
Alibaba, Tencent, ByteDance and other Chinese AI developers have ordered about 6 million Nvidia RTX Pro 5500 chips worth $70B, Bloomberg reports. Sources say they expect to receive only around 1% of those chips within the next four months, highlighting a large gap between demand and supply for advanced Nvidia hardware in China.
1 source
Techmeme
Anthropic
Developing · since Tue
Anthropic reports progress on FlashProve, a way to prove which AI model produced an output
Anthropic shared progress on FlashProve, an open-source system for provable inference that cryptographically verifies an AI response came from the model it claims to be from. Model fingerprints allow verification without revealing weights. The work could prevent labs or routers from quietly swapping models, though there are no deployment plans yet.
1 new · 3 sources
Anthropic
TechCrunch
Techmeme
More from the day
AI industry
10:55
Tencent releases Hy4 open-weight foundation model
Tencent has released Hy4, an open-weight foundation model the company says reaches the top tier of Chinese open-weight AI for the first time. A faster Flash version is planned. The release puts Tencent alongside China's leading open-model labs.
1 source
Techmeme
AI industry
19:45
AMD to pay about $2.9 billion for AI chip startup Taalas
A regulatory filing shows AMD will pay roughly $2.9 billion in cash and stock for Toronto-based AI chip startup Taalas, with about $1.4 billion of that contingent on unspecified conditions. The acquisition was first announced in August, and the filing now details the full deal value.
1 source
Techmeme
AI industry
07:20
Chinese embodied AI startups raised $13B in H1 2026, up 5x
Per ITjuzi data, Chinese embodied AI companies raised roughly $13 billion across 322 deals in the first half of 2026, a fivefold year-over-year increase. Robot makers are shifting focus from hardware feats like backflips toward AI capabilities, fueling the funding surge.
1 source
Techmeme
Anthropic
15:45
Google backs TeraWulf data center project tied to Anthropic
TeraWulf said it secured a credit backstop agreement with Google to support financing for a Kentucky data center project involving Anthropic. The facility is expected to come online in the second half of 2027. The deal extends Google's financial backing of Anthropic's computing infrastructure.
1 source
Techmeme
Nvidia
08:46
Nvidia RTX Spark encrypted firmware severely degrades AI inference speed on Linux
Encrypted firmware within the Nvidia RTX Spark is causing significant computational slowdown. Testing on Linux platforms showed inference speeds dropped to just 35%. This issue impacts AI developers relying on these hardware tools for efficient deployment.
1 source
mindstudio.ai
Policy & law
00:42
Nexus Mods mandates AI usage disclosure for new uploads
The PC gaming mod repository Nexus Mods is requiring creators to declare if they utilized artificial intelligence in their work. This new policy aims to address the growing volume of AI-generated content on the platform.
1 source
The Verge
OpenAI
00:40
Cat Wu joins OpenAI to lead product management for core products
Cat Wu, who left Anthropic last week after leading Claude Code, says she is joining OpenAI to lead product management for core products, announcing the move on X. Her exit follows Thariq Shihipar's departure last month, and she is now taking on a senior product role at a top rival.
1 source
Techmeme
Nvidia
00:20
Report: Nvidia wields soft power over AI customers and US chip policy
The Information reports Nvidia uses influence over customers to shape US chip policy, including lobbying and relationships with AI labs. Sources said OpenAI avoided taking a public stance on China chip exports to avoid angering Nvidia. The story highlights Nvidia's growing political and commercial leverage in the AI ecosystem.
1 source
Techmeme
Google
12:00
Google search growth slows as AI monetization lags
Alphabet reported that Google search queries are growing at their fastest rate in five years, but search revenue growth slowed to 12% year over year in Q3, down from about 17% the prior quarter. The gap suggests Google has not yet found a way to monetize the surge in AI-driven queries as effectively as traditional search ads.
1 source
The Verge
OpenAI
07:00
OpenAI's Daybreak Red lands in AWS GovCloud via Bedrock
OpenAI announced that Daybreak Red is now available in AWS GovCloud (US-East) through Amazon Bedrock. The release brings GPT-5.6 Cyber to approved customers for authorized cybersecurity work. It matters because it extends OpenAI's specialized cyber model to regulated government cloud environments.
1 source
OpenAI
OpenAI
21:25
OpenAI's Chris Lehane pitches AI's role in the 2028 election
OpenAI's Chris Lehane is reportedly proposing that AI should shape the 2028 US election, framing it as a new social contract for Americans. The effort could also advance OpenAI's own business interests, drawing attention to the company's growing political ambitions.
1 source
TechCrunch
Google
11:05
Google expands voluntary buyouts to five more countries amid AI spending
Google expanded its voluntary exit buyout program in June to employees in Canada, France, Germany, Japan, and Switzerland, according to Reuters. The move reflects Alphabet's effort to rein in costs as it spends heavily on AI infrastructure, and signals continued cost discipline across its international offices.
1 source
Techmeme
Policy & law
14:00
Thinktank warns Australia risks another robodebt over AI in government decisions
The Centre for Public Integrity has warned the Albanese government it risks a repeat of the robodebt scandal due to weak transparency around government use of algorithms and AI. A new report calls for mandatory auditing and a public register of automated decision-making.
1 source
The Guardian
OpenAI
00:00
DevRev builds autonomous service tools on GPT-6.1 Sol
DevRev announced it is using OpenAI's GPT-6.1 Sol to power its Computer and Operator products, which provide autonomous self-service for enterprises. The company says the model balances reasoning, speed, and cost, showing how frontier models are being adopted for customer support automation.
1 source
OpenAI
AI industry
17:02
Pangram launches Image 1.0 and Image Flash 1.0 detection models
Pangram released Pangram Image 1.0, which it says is state-of-the-art for detecting AI-generated images, alongside Image Flash 1.0, which powers its AI video detection. The launch moves the tools out of a research preview several months after they first appeared. It matters as demand grows for tools identifying synthetic media.
1 source
Techmeme
Policy & law
23:30
Data center stocks slide on expectations of slower AI buildout
Data center-related shares fell after Democrats won control of Congress in the midterm elections, with Hut 8 dropping 23%, IREN 11%, Core Scientific 10% and Applied Digital 7%. Investors expect the election outcome to slow the pace of AI infrastructure buildout, even as the broader market closed sharply higher.
1 source
Techmeme
AI industry
04:20
Pharma leaders doubt AI can cure most diseases within a decade
Drug industry executives are pushing back on claims that AI could cure most diseases within ten years, the Wall Street Journal reports. No AI-designed biotech drug has yet won FDA clearance, and leaders argue biology itself remains the main obstacle to AI-driven drug discovery.
1 source
Techmeme
AI industry
13:01
Nooks secures $500M commitment from General Catalyst's CVF
AI sales automation startup Nooks has secured a $500 million commitment from General Catalyst's Customer Value Fund. The capital will finance Nooks' sales and marketing push, with General Catalyst taking a share of the returns generated by the capital rather than traditional equity.
1 source
Techmeme
AI industry
18:44
Hugging Face policy chief on open source AI's economics and risks
TechCrunch published two interviews with Hugging Face's policy chief discussing the rise of open source AI models. Enterprises, governments and startups are adopting open models for lower costs, control and flexibility, but questions remain over how companies capture value and whether American players can keep pace with Chinese open models.
1 source
TechCrunch
AI industry
20:10
Apple cuts internal AI usage budgets for engineers
Apple has reduced daily AI spending caps for some engineers from $300 to $200, with plans to cut further to as little as $100, Bloomberg reports. The cost-cutting move has stoked staff concerns about possible layoffs. It signals tighter scrutiny of internal AI tool expenses across the company.
2 sources
TechCrunch
Techmeme
Google
20:23
AT&T smart security integrates Gemini for Home
AT&T says its Connected Life smart home security system is the first to integrate with Google Home Premium, bringing Gemini-powered features to its security offering. Users can get summaries of home events and search their video history. The move extends Gemini for Home into third-party security products.
1 source
The Verge
AI industry
16:51
Hands-on: Apple's Siri Recaps prove useful, if a bit unsettling
A TechCrunch reporter spent a day testing Apple's AI-powered Siri Recaps and found them surprisingly useful, despite initial privacy concerns. The review suggests the feature won the writer over quickly, though improvements are still needed.
1 source
TechCrunch
AI industry
00:14
Canva to cut or relocate over 1,200 jobs in major restructure
Canva is cutting or relocating more than 1,200 roles, about 22% of its global workforce, in a major restructure. CEO Melanie Perkins called it the hardest moment in the company's thirteen-year history. The move reflects a broader trend of software companies restructuring across the tech sector.
1 source
The Guardian
AI industry
18:53
Ricursive previews Ripple-65B open-weight model for agentic coding
Ricursive announced a preview of Ripple-65B, an open-weight model it says achieves US open-weight state-of-the-art performance on agentic coding tasks. Details are limited to the company's own claims, and independent benchmarks have not yet verified the results. If confirmed, it would strengthen the open-weight ecosystem for coding agents.
1 source
ricursive.com
AI industry
18:13
Yale researchers find AI chatbots can skew election information
A Yale study had people use AI chatbots to look up election information and found mixed results. Researchers say chatbots can skew election-related answers and act as an intervening layer in voters' information searches. The findings highlight reliability concerns as more voters turn to AI tools.
1 source
ZDNET
AI industry
13:15
Sentra raises $150M Series C for AI agent data security
Data security startup Sentra raised a $150 million Series C led by Capital One Ventures, bringing its total funding above $250 million. The company builds tools that restrict and monitor data access for AI agents, and will use the new capital to expand its platform.
1 source
Techmeme
AI industry
19:16
Hugging Face CEO says his AI doom probability is close to zero
Hugging Face CEO Clément Delangue said his estimate of existential risk from AI is close to zero, arguing that risk mostly comes from a few companies building behind closed doors. He framed the remarks as support for open source AI, saying open development reduces concentrated risk.
1 source
Wired
AI industry
13:00
MIT Tech Review urges data lineage to secure AI agents
MIT Technology Review describes how AI agents can chain errors across enterprise workflows, using a sales example where an agent pulls restricted pricing data and a second agent drafts a proposal with no human checks. The piece argues data lineage tools are needed to secure such agentic pipelines.
1 source
MIT Technology Review
Google
18:07
Google Research explores AI-assisted agile deliberation for gray areas
Google Research published a piece describing an agile deliberation approach where AI helps humans work through ambiguous decision areas. The post outlines how AI tools could support group discussions and judgment calls where rules are unclear. Its significance lies in framing AI as an aid to human deliberation rather than automated decision-making.
1 source
Google Research
Policy & law
23:00
Victorian parties trade accusations over AI-generated election videos
Victoria's major political parties have used AI-generated videos attacking each other's leaders in state election campaign ads. Transparency and advertising experts warned that such deepfakes could further erode public trust in politics.
1 source
The Guardian
AI industry
17:14
Apple beta brings deeper Siri AI and Safari customization
A new Safari developer beta includes deeper Siri AI integration and expanded browser customization options, ZDNET reports. The release notes suggest Apple is continuing to weave AI features into its core products. The changes remain in beta and are not yet generally available.
1 source
ZDNET
AI industry
11:00
Robot dogs move from shopping helpers to security roles
A Guardian feature examines the growing use of four-legged robot dogs, from a San Francisco startup paying walkers $20 an hour to demo them, to US immigration agency ICE buying units and a startup called Warhound planning combat variants. It highlights how consumer novelty is giving way to policing and military applications.
1 source
The Guardian
AI industry
14:00
Australians rank AI ahead of climate as top future concern
The annual Scanlon report on social cohesion finds most Australians support immigration and multiculturalism, but cost of living and artificial intelligence now outrank climate change and war as concerns for the future. The survey also records growing distrust of government among Australians.
1 source
The Guardian
AI industry
18:02
Deployment engineers among top AI earners, study finds
A new study covered by ZDNET reports that AI deployment engineers are earning salaries as high as $1.45 million a year, comparable to executive-level pay. The finding highlights how new AI-focused roles command premium compensation. It underscores the demand for specialists who put AI systems into production.
1 source
ZDNET
AI industry
00:04
Graduates increasingly spend early careers managing AI tools
A BBC report says recent graduates are spending more of their work time overseeing AI systems. The piece raises concerns that heavy AI management may crowd out broader early-career experience. It reflects growing debate over how AI is reshaping entry-level jobs.
1 source
BBC News
AI industry
09:45
Engineer reports 10x PR output at AI-native startup Tano
Software engineer Neil Kakkar writes that his monthly merged pull requests rose from eight to 20 before joining, to more than 200 after eight months at Tano, an AI-native creator startup. The account offers a firsthand view of how AI tooling and company practices can shape developer productivity, though the figures are self-reported and unverified.
1 source
Techmeme
OpenAI
14:46
Sam Altman defends letting people put as much RAM as they want in PCs
OpenAI CEO Sam Altman spent election night responding on X to comments about PC memory, changing his display name to Ram Altman. He said he wants people to be able to put as much memory as they want in their computers. The exchange was a lighthearted social media moment, not a company announcement.
1 source
The Verge
AI industry
Developing · since Tue
Netflix rolls out AI Discover tab for mood-based viewing picks
Netflix has begun gradually rolling out a new AI-powered Discover tab on its US TV app for English-language users. The first experiment suggests shows by theme such as Something Light or Family Movie Night, and learns when users reject recommendations. It aims to counter falling viewing time announced alongside recent layoffs.
1 new · 3 sources
ZDNET
TechCrunch
The Verge
Policy & law
Developing · since Tue
NYC mayor Zohran Mamdani details municipal AI policy views
New York City mayor Zohran Mamdani gave a Wired interview ahead of the midterms laying out his positions on AI. He supports banning AI use for some students and backs a one-year pause on data center development. The interview signals how city leaders may shape AI and data center policy.
1 new · 2 sources
Techmeme
Wired
AI industry
Developing · since Tue
Hugging Face releases Dharma Field Extractor after StateBench debut
Hugging Face launched Dharma Field Extractor, a tool that turns any document into structured JSON fields. It follows the company's StateBench benchmark, released a day earlier, which tests how well language models track state changes across long documents.
1 new · 1 source
Hugging Face