Full archive

Latest AI News

Search reporting on AI markets, models, products, companies, policy, and culture.

Showing 1-18 of 173Results matching "Shift"

Today

7 stories

1 hour ago

GPT-6 Astra Hits 99.9% on ARC-AGI-3, but Scores 62.7% in a Shared Test

ARC Prize will label the two evaluation conditions separately after OpenAI’s context-management setup produced a near-perfect result that its shared interface did not reproduce. GPT-6 Astra reached 99.9% on ARC-AGI-3 ’s Semi-Private set at high reasoning effort through a Provider Adapter harness. The benchmark places agents in unfamiliar abstract, turn-based environments without explicit instructions.

3 min readRead story

Lawsuit Policy

2 hours ago

Meta Lawsuit Adds Bystanders Who Say AI Glasses Recorded Them Without Consent

The amended complaint shifts the dispute from what Meta told customers to whether people near a glasses wearer had any meaningful say over footage that plaintiffs allege entered an AI-review and training process. Meta ’s AI-glasses lawsuit now includes people who never bought the product. The case is in the Northern District of California, and no merits ruling has been issued.

3 min readRead story

Business

2 hours ago

Robot Startups Turn to Human Workers as Simulation Hits Manipulation Limits

Companies are recording everyday work, teleoperating machines and capturing motion with wearables. Simulation is scaling quickly, but its toughest transfer problems leave real-world examples in demand. Robot makers are trying to manufacture the ingredient that language models found on the internet: a huge supply of examples.

3 min readRead story

Products

3 hours ago

xAI Explains How Grok Bot Is Built to Keep Agents Working Between Chats

xAI’s design sets a Bot—not a disposable conversation—as the durable unit of work. The harder test is whether users can grant agents enough autonomy while retaining clear ways to inspect, redirect, and stop them. xAI is describing Grok Bot as an agent workspace built for work that survives the chat that started it.

4 min readRead story

Launch Models

7 hours ago

Google Puts WeatherNext 3’s Hourly AI Forecasts Into Search, Maps and Gemini

The launch moves an observation-led weather model from Google’s research stack into consumer products and cloud data tools. Its performance claims are substantial, while Google directs users seeking official severe-weather warnings and public-safety advisories to meteorological agencies. That variable-by-variable design means the headline 5-kilometer figure does not apply to every output.

2 min readRead story

Viral Culture

10 hours ago

Restaurants Turn to AI Food Images, and Diners Are Pushing Back

Cheap synthetic images are drawing criticism, while restaurant software companies pitch AI for menu research and improvements to photos of real dishes. AI-generated food images are appearing on delivery menus and restaurant posters, where consumers are criticizing their uncanny and unappetizing appearance. Restaurant marketing changed as more customers began choosing meals through online menus and delivery apps.

2 min readRead story

Partnership Culture

12 hours ago

Tinder Begins Requiring Face Checks for Existing Users in the US and UK

The expansion turns a tool introduced for new accounts into a broader condition of participation, while leaving a central limit intact: a biometric check can show someone appeared at sign-up, not that a profile is trustworthy. Dating-app users in the US and UK are beginning to face a new condition for staying on Tinder: a face check. The practical aim is to make fake accounts, scams, and spam harder to run at scale.

3 min readRead story

Yesterday

11 stories

Launch Products

16 hours ago

Uber and Wayve Put 15 Supervised Robotaxis on London’s Streets

The launch brings Wayve’s AI driving system into paid passenger service in London, but the fleet remains small and a safety driver remains responsible for every trip. Uber riders in London can now be matched with Wayve autonomous vehicles through the app. The service starts with 15 licensed cars, each carrying a licensed human driver who can take control if needed.

2 min readRead story

Launch Business

19 hours ago

Meta Drops Token Counts From Reviews While Employees Test Hatch

Meta is separating performance assessment from AI activity just as it puts a more autonomous tool in front of employees. Whether workers trust Hatch with personal accounts remains a separate test. Meta has told employees that performance reviews will no longer depend on how much they use AI tools.

2 min readRead story

Platform shift Infrastructure

21 hours ago

Broadcom Forecasts $230 Billion in AI Chip Revenue for Fiscal 2028

The long-range target would put Broadcom on a far larger AI-chip trajectory. Its fourth-quarter guidance is the next concrete measure of whether that momentum continues. Broadcom CEO Hock Tan forecast that the company’s AI-chip revenue will reach about $115 billion in fiscal 2027 and $230 billion in fiscal 2028.

2 min readRead story

Earnings Business

Yesterday

HPE Raises 2026 Revenue Growth Forecast to 34%-37% After $12.21B Quarter

The earnings beat gives HPE a stronger base for its new forecasts, while constrained components could limit how quickly server demand becomes delivered systems. Hewlett Packard Enterprise lifted its fiscal 2026 and 2027 growth forecasts after a quarter in which AI server and networking demand helped revenue and adjusted earnings exceed analyst expectations.

2 min readRead story

Enterprise adoption Business

Yesterday

Napster Pursues Penny-a-Minute AI Agents Through Microsoft

The former music platform is betting that a low price can make enterprise agents easier to sell. A 2025 coaching pilot shows the Microsoft relationship has already moved beyond a partnership announcement. Napster is pursuing low-cost AI agents through Microsoft, putting a price of about a penny per minute at the center of its next business direction.

2 min readRead story

Government action Policy

Yesterday

New Mexico Lets Districts Seek Approval to Leave K-2 AI Reading Tests

The 2026-27 opt-out route does not end statewide use of Amira. It shifts a disputed early-literacy system toward district-by-district decisions on assessment, student recordings and oversight. The change is limited: a district must receive approval from the New Mexico Public Education Department to leave the program.

2 min readRead story

Security risk Models

Yesterday

OpenAI Says Astra Will Use Chain-of-Thought Monitoring Despite Reported Design Concern

OpenAI is preparing to restrict access to its most capable cyber features while relying on monitors that inspect reasoning and actions. But a reported recurrent-depth design has raised a separate question: how much of the model’s work will remain visible as text. OpenAI says Astra can find previously unknown security flaws and develop exploits against well-protected systems with the right tools and access.

3 min readRead story

Acquisition Business

Yesterday

LivePerson Stockholders Approve SoundHound AI Deal, Targeting a September 4 Close

The vote advances the proposed conversational-AI combination, while a notes restructuring and other closing conditions remain part of the path to completion. LivePerson stockholders approved SoundHound AI’s proposed acquisition at a September 2 special meeting, putting the companies on track for an expected September 4 close. Completion remains subject to customary closing conditions.

2 min readRead story

Funding Startups

Yesterday

HiddenLayer Raises $100M for Runtime Controls on Enterprise AI Agents

The startup is funding a security layer meant to observe and intervene as agents use enterprise tools. Its test is whether buyers will keep purchasing that layer independently of the platforms hosting their AI. HiddenLayer has raised a $100 million Series B to expand security software for enterprise AI systems, with a particular focus on watching AI agents while they operate.

2 min readRead story

Launch Products

Yesterday

AWS Makes Its Contact-Center AI Designer Generally Available With Rules Beside the Model

The product separates flexible conversation from fixed decisions such as identity checks and eligibility, while linking voice calls to on-screen choices and preserving context during a transfer. AWS has made agentic CX designer generally available in Amazon Connect Customer, giving contact-center teams a no-code canvas that places generative AI conversation steps alongside fixed business-rule steps.

3 min readRead story

Government action Policy

Yesterday

New York City Public Schools Will Block Student AI Use Through Eighth Grade

The planned rules would remove generative AI, AI tutors and other student-facing AI software from preschool through middle school, shifting the district’s focus toward supervised AI literacy in high school. Enforcement beyond school devices remains an open problem. The policy is scheduled for announcement Sept.

3 min readRead story

Load more stories18 more next