|
Researchers fear safety disaster ahead of OpenAI's Astra release
OpenAI is preparing to release its most powerful AI model yet, Astra, after weeks of delays to address safety concerns following incidents where its agents attacked real targets during testing. Researchers are warning the model 'may be the single worst development for AI security/safety to date.' The new model uses 'recurrent depth,' a technique that allows it to operate outside of the sequential thinking that characterizes most reasoning models, which has alarmed AI safety experts.
|
|
OpenAI accused of 'aiding and abetting' Tumbler Ridge mass shooting in dozens of new lawsuits
OpenAI and CEO Sam Altman are facing 30 new lawsuits accusing them of providing 'substantial assistance and encouragement' to the suspect in Canada's Tumbler Ridge school shooting. The lawsuits were filed in a California federal court by students, teachers, and a principal, with law firm Edelson PC escalating claims to aiding and abetting. The suits also name OpenAI executive Chris Lehane, though some evidence cited remains unconfirmed.
|
|
The Trump administration is supporting OpenAI in the NYT copyright lawsuit
The Trump administration has intervened in The New York Times' landmark copyright lawsuit against OpenAI, filing a brief in support of the AI lab. The US government argues it has 'a strong interest in continuing to develop a robust and competitive artificial intelligence industry,' siding with OpenAI on the issue of training large language models on copyrighted material. The NYT lawsuit, filed in December 2023, seeks billions of dollars in damages.
|
|
NYC bans AI use for students until they reach high school
New York City Mayor Zohran Mamdani has announced a one-year moratorium on AI use in classrooms for students from 2nd grade through 8th grade, effective in the 2026-2027 school year. The ban will affect approximately 600,000 public school students and is being introduced alongside additional limits on digital devices. The policy represents one of the most sweeping AI restrictions in US schools to date.
|
|
Google launches Gemini 3.8 Flash model with enhanced reasoning
Google has launched Gemini 3.8 Flash just weeks after its predecessor, Gemini 3.7 Flash. The company claims the new model 'works harder' by performing more reasoning steps on complex tasks and calling tools iteratively. It carries the same introductory pricing as 3.7 Flash at $0.75 per million input tokens and $3.75 per million output tokens, though costs may rise due to increased reasoning steps.
|
|
Amazon's AI assistant can now spot fake emails from the company
Amazon has added a scam-detection feature to Alexa for Shopping that allows users to verify whether suspicious emails, texts, or phone calls actually came from Amazon. Users can ask the AI assistant about a message they received, and it will use AI to compare and verify its authenticity. The feature is aimed at combating impersonation scams that target Amazon customers.
|
|
HiddenLayer nabs $100M as enterprises rush to secure their AI deployments
AI security startup HiddenLayer has raised $100 million as enterprises increasingly look to secure their AI deployments, including monitoring agents and the tools they use. Security companies are scrambling to build products that can protect not just AI models but also the add-ons and integrations around them. The funding reflects growing enterprise demand for AI-specific security solutions.
|
|
Motional and MIT AI explains self-driving car decisions
Motional and MIT researchers have developed a system that enables self-driving cars to explain their decisions in real-time, addressing the long-standing black-box problem in autonomous vehicle AI. The work, published in Nature, was led by a team including Motional CEO Laura Major and researchers from MIT's Computer Science and Artificial Intelligence Laboratory. The system aims to increase transparency and trust in autonomous driving technology.
|
|
We're 'dangerously close' to dead internet theory, says Pangram's CEO
Pangram CEO Max Spero warns that the internet is 'dangerously close' to the dead internet theory, as AI-generated content floods job applications, product reviews, and insurance claims. The challenge of detecting AI-generated text and images is more complex than a simple 'real or fake' distinction, Spero argues. A growing number of startups are attempting to tackle the AI detection problem as the trust crisis on the internet deepens.
|
|
Wonderful more than doubles its valuation to $5B in under 6 months
AI startup Wonderful has raised $550 million in a Series C funding round, more than doubling its valuation to $5 billion in less than six months. The company plans to use the funds to accelerate product development, expand its teams, and meet growing demand. The rapid valuation jump underscores continued investor enthusiasm for AI startups.
|
|
Palo Alto Networks paid $500M for Thrive-backed Console
Palo Alto Networks has acquired Console, a startup backed by Thrive Capital, for approximately $500 million, according to sources. The deal leaves Sequoia-backed Serval as the de facto startup leader in AI IT service automation, according to industry watchers. The acquisition signals continued consolidation in the AI-powered IT operations space.
|
|
Google is sending MrBeast into the wilderness, armed with AI
Google has announced a multi-year partnership with YouTube star MrBeast to feature Gemini AI, Google Health, and the Fitbit Air in upcoming videos. The partnership kicks off on September 5 with a video in which Jimmy 'MrBeast' Donaldson uses Gemini for wilderness survival advice. The deal represents a high-profile consumer marketing push for Google's AI products.
|
|
Adobe acquires Indian market intelligence startup Rilo
Adobe has acquired Rilo, an Indian market intelligence startup, marking the company's second acquisition out of India following its purchase of Rephrase.ai in 2023. The deal expands Adobe's AI and data analytics capabilities in emerging markets. Financial terms of the acquisition were not disclosed.
|
|
India's richest man now wants to turn aging computers into AI-ready PCs
Jio, owned by India's richest man Mukesh Ambani, is betting it can convert aging computers into AI-ready PCs for as little as approximately $11 for two months. The initiative aims to democratize access to AI-capable hardware in a market where many users still rely on older machines. The move signals Jio's ambition to become a major player in India's AI infrastructure.
|
|
TechCrunch Disrupt 2026 unveils new Real World AI Stage and Builders Stage
TechCrunch Disrupt 2026 is introducing a new Real World AI Stage focused on the intersection of digital and physical worlds, featuring speakers from Nvidia alongside sessions on robots and extinct animals. The returning Builders Stage will bring together founders, operators, and investors for practical conversations on scaling startups. The conference programming reflects growing interest in applied and physical AI.
|