Page 6 - Ai Models

Meta AI model accessed internet, hacked outside firm during testing

The company said a testing misconfiguration gave its AI model internet access, allowing it to exploit a vulnerability in an external service during a cybersecurity evaluation

OpenAI, Anthropic model implicated in new security breaches during tests

Britain's AI Security Institute said agents powered by Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol engaged in unauthorized actions during security evaluations

Apple seeks preliminary injunction against OpenAI in trade secrets case

Apple also filed a concurrent motion seeking expedited discovery, including ​production of documents relating to the defendants' alleged access of Apple's proprietary and trade secret information

China's AI advance creates 'death zone' for rival US model makers

A wave of advanced AI models from Alibaba, Moonshot, ByteDance, DeepSeek and Z.ai is rapidly closing the technology gap with US rivals while intensifying competition on price and performance

Meta, Anthropic, OpenAI to meet Trump officials about AI safety testing

Anthropic and OpenAI disclosed in recent days that their AI tools breached the systems of other companies, stirring concerns among US ‌lawmakers

EU in talks with OpenAI, Anthropic after rogue AI agent hacking incidents

Europe's AI Act, which kicks in on August 2, is the first legislation in ‌the world to regulate a technology ​used in almost all sectors of ​businesses and society

DeepSeek's V4-Flash emerges as world's cheapest AI model to run: Report

DeepSeek officially released ​its V4-Flash model on Friday, its latest ​attempt to regain momentum by doing what it is best known for - offering ultra-low-cost AI alternatives

Alibaba unveils largest AI model Qwen3.8-Max as competition intensifies

Qwen3.8-Max has 2.4 trillion parameters, the numerical settings ​a model learns from data and uses to recognise patterns, generate answers, and carry out tasks

Anthropic, OpenAI cyber failures raise concerns over US national security

Cybersecurity experts criticised Anthropic and OpenAI after AI models accessed external organisations during testing, with Anthropic's Claude stealing credentials and data due to a configuration error

Anthropic AI models hacked 3 organizations during cybersecurity tests

Anthropic said in a blog post Thursday that it made the discovery after performing a review of its own cybersecurity tests, following OpenAI's announcement of a breach

Sarvam ropes in Mistral founding team member Devendra Chaplot as adviser

Sovereign AI startup appoints former Mistral and xAI executive as adviser as it expands globally and builds a one trillion-parameter foundation model

ADA acquires Algonomy, strengthening its intelligent growth platform for a fully agentic experience

.

Police probe alleged AI use by protesters to counter crowd-control measures

Investigation into the July 20 clashes during the "Chalo Sansad" protest organised by the CJP has revealed that some protesters may have used AI tools to learn methods of countering police crowd-control measures, including use of wet jute bags to minimise effects of tear gas, police sources said on Thursday. According to the source, the ongoing investigation has indicated that some participants may have searched artificial intelligence (AI) platforms for information on mitigating the impact of tear gas, and preparing for confrontations with security personnel. "The investigation is being carried out through multiple methods. We have reasons to believe that some protesters used AI tools or other generative AI platforms to search for ways to immediately reduce the effects of tear gas, including use of wet jute bags, and to prepare themselves with hard hats, masks and goggles," the source told PTI. He said police are in the process of collecting evidence, which is a time taking ...

EU AI Act: What changes as new transparency rules take effect from August 2

The European Union's AI transparency rules become applicable from August 2, bringing mandatory disclosures for certain AI-generated content and significant penalties for violations

Why over 1000 AI researchers want governments ready to slow AI progress

More than 1,100 AI researchers from OpenAI, Anthropic, Google and Meta have urged the US government to prepare mechanisms to deliberately pace frontier AI development. Here is the reason

Why Mark Zuckerberg thinks banning Chinese AI models won't help the US

Meta CEO Mark Zuckerberg said blocking advanced Chinese AI models in the US would not be an effective response to growing competition from China

US govt should not block Chinese models to gain edge in AI race: Zuckerberg

The Trump administration ‌on Tuesday unveiled bans ​that target imports of ‌new Chinese robots and power inverters, seeking to protect ‌the US AI buildout from national security threats

More than 1,100 workers of AI firms calls on US govt to pace tech growth

The letter has been signed by employees across nearly a dozen firms, including OpenAI, Anthropic PBC, Alphabet Inc.'s Google and Meta Platforms Inc

OpenAI's AI agents compromised a customer at second tech firm Modal

The agent broke into an isolated testing environment known as a sandbox that Modal was running for a customer, Akshat Bubna, chief technology officer of the cloud platform for developers said

Open AI models gain industry support, but security commitments diverge

Google and OpenAI supported Nvidia's policy case for open-weight models but did not join its security alliance, while Anthropic challenged claims that openness inherently favours defenders