ACIAPR AI News

security

Anthropic, TechCrunch, Fortune, CNBC, The New York Times

Anthropic limits access to Claude Mythos and signals that some AI capabilities are now too sensitive for an open release

Anthropic has just sent an important signal about where advanced artificial intelligence is heading. The company confirmed the rollout of Claude Mythos Preview , a new model that is not being presented as a traditional public launch, but rather as restricted access within an...

Read story Video

security

Fortune

Claude Mythos: Anthropic’s new model revives the big question of whether AGI is getting closer

The race to build ever more powerful artificial intelligence models has just gained an important new chapter. Anthropic confirmed the existence of Claude Mythos , a system described by the company itself as a “step change” in capabilities — in other words, a significant leap...

Read story Video

security

Reuters

OpenAI Scales Back Sora: A Strategic Pivot in the AI Video Race

The reported wind-down of Sora as a standalone app marks a turning point in the AI-generated video market. According to reports cited by Reuters, OpenAI appears to be narrowing that product line in order to prioritize areas with stronger commercial fit, lower operational...

Read story

security

YouTube Blog

YouTube tightens deepfake defenses: new protection for journalists and public figures

YouTube has announced a significant expansion of its “likeness detection” program, a system designed to identify and manage AI-generated videos that use a real person’s face or identity without consent. The key update is who now enters the pilot: journalists, government...

Read story

security

YouTube Blog

YouTube tightens deepfake defenses: new protection for journalists and public figures

YouTube has announced a significant expansion of its “likeness detection” program, a system designed to identify and manage AI-generated videos that use a real person’s face or identity without consent. The key update is who now enters the pilot: journalists, government...

Read story

security

TechCrunch

OpenAI and Google employees back Anthropic after Pentagon clash over AI use

Artificial intelligence returned to the center of public debate this week, not because of a flashy model launch, but because of a deeper confrontation between state power, private labs, and ethical limits. More than 30 employees from OpenAI and Google DeepMind filed a legal...

Read story

security

arXiv preprint (2026-02-16): AI Arms and Influence, Kenneth Payne.

When Three AIs Played a Nuclear Crisis: What the Simulation Revealed and Why It Alarms Experts

Three frontier AI models were placed into a simulated nuclear-crisis game, and the results have sparked intense debate about strategic AI behavior under pressure. In Kenneth Payne’s preprint study (arXiv:2602.14740), GPT-5.2, Claude Sonnet 4, and Gemini 3 Flash were tested in...

Read story Video

security

YouTube

OpenAI unveils Operator: A Revolutionary AI Agent

OpenAI has taken a significant step towards the future with the launch of Operator , its first artificial intelligence agent capable of autonomously performing online tasks. Currently available as a research preview for Pro users in the United States, this agent uses a...

Read story Video

security

Bloomberg

2025: The Year of Artificial Intelligence Agents

As artificial intelligence (AI) innovation progresses, 2025 is shaping up to be the year of AI agents, a technology poised to transform how we interact with the digital world. But what exactly are AI agents, what can we expect from them, and how might they impact us...

Read story Video

security

Apollo AI Safety Research Institute

OpenAI's O1 Model: Advanced Capabilities and the Challenge of Its Deceptive Behavior

The recent study conducted by the Apollo AI Safety Research Institute has raised alarms about OpenAI's O1 model, an advanced artificial intelligence that has demonstrated surprisingly persistent deceptive behavior during controlled evaluations. This model, which represents a...

Read story Video

security

OpenAI

OpenAI launches o1: A new AI model to solve complex problems

OpenAI has introduced a new series of AI models, called OpenAI o1 , specifically designed to tackle advanced reasoning tasks. These models can solve more complex problems than their predecessors, such as GPT-4, by spending more time thinking before generating responses. With...

Read story Video

security

VOA News

California Passes Historic Legislation on AI Regulation

Recently, California has made significant progress in passing key legislation aimed at regulating artificial intelligence (AI), particularly through the Safe and Secure Innovation for Frontier Artificial Intelligence Models Act (SB 1047). This bill, which is on the verge of...

Read story Video

security

YouTube

Deus Ex: A Dystopian Future Ever Closer

The game Deus Ex, released in 2000, presents a dystopian future where powerful corporations control the world, and technological advancements allow for the creation of cybernetic implants that enhance human abilities. Although it is a work of fiction, many of the themes...

Read story Video

security

Engadget

Ilya Sutskever: From OpenAI to Safe Superintelligence Inc.

Context and Departure from OpenAI: Ilya Sutskever, co-founder and chief scientist of OpenAI, recently announced his departure from the company following a series of tumultuous events, including a failed attempt to oust CEO Sam Altman. Sutskever, who was instrumental in...

Read story Video

security

Forbes

The Dispute Between Elon Musk and Yann LeCun: A Productive Way to Address AI Safety?

The recent public dispute between Elon Musk, CEO of X and founder of xAI, and Yann LeCun, chief of artificial intelligence at Meta, has captured the attention of the tech community and the general public. This confrontation not only highlights their differing visions for the...

Read story