Follow on Telegram

ai/models

Researchers Expose Memory Poisoning as New Attack Vector for AI Agents

As AI agents gain the ability to remember past interactions across sessions, a new cybersecurity threat has emerged: memory poisoning attacks. AI agents are autonomous systems that use AI models to plan and execute tasks on behalf of users. A growing number of these agents now incorporate persistent memory—allowing

AI Safety Debate: Companies Explore Models with Minimal Restrictions

The Push for Unrestricted AI Models The artificial intelligence industry is witnessing a significant shift as some companies explore building AI models with minimal safeguards. Unlike mainstream AI assistants that refuse certain requests, these new models aim to process inputs without default "no" responses. What This Means for

DOJ Backs OpenAI and Microsoft in New York Times Copyright Lawsuit

The U.S. Department of Justice has stepped into a high-profile copyright dispute between the New York Times and AI companies OpenAI and Microsoft, filing a statement of interest that urges a federal judge to rule in favor of the technology defendants. The New York Times lawsuit, originally filed in

Technology News

OpenAI's Recurrent Depth Approach Sparks Safety Concerns

OpenAI is working on a new reasoning approach called "recurrent depth" that could fundamentally change how AI models process information. The technique, intended for the upcoming Astra model, allows the system to operate outside the step-by-step chain-of-thought reasoning that currently characterizes most advanced language models. Unlike traditional reasoning

Trump Administration Backs OpenAI in New York Times Copyright Lawsuit

The Trump administration has intervened in the ongoing copyright dispute between OpenAI and the New York Times, filing a legal brief in support of the AI company. The Department of Justice submitted the filing on behalf of OpenAI, arguing positions that diverge from the newspaper's claims that the

Technology News

US Government Files Brief Supporting OpenAI's Position on Training Data Copyright

The United States government has weighed in on the ongoing legal debate over AI training practices, filing a brief that supports OpenAI's position on using copyrighted material to train large language models. The brief emphasizes the country's "strong interest in continuing to develop a robust

Technology News

Trump Administration Backs OpenAI in New York Times Copyright Lawsuit

The Trump administration has entered the legal battle between OpenAI and The New York Times, throwing its support behind the AI company in a case that could reshape how the technology industry trains large language models. In a filing submitted this week, the administration argued that The New York Times