Page 2 - Ai Models

Anthropic, OpenAI sound AI safety alarm, seek to shape how it's controlled

As tech titans warned an AI-weary world that their own advanced systems could endanger humanity, the question emerged: What do they have to gain? The answer may lie in the companies' efforts to score political goals ahead of the US midterm elections and the money top artificial intelligence labs and their investors stand to make in much-anticipated listings on a jittery stock market. The CEOs of Anthropic and OpenAI recently declared America's cutting-edge models are so powerful, they're dangerous, and need to be regulated and independently tested before being released. In a rare instance of unity, they have sketched alarming scenarios in essays, social media posts and speeches to the United Nations. Along the way, they are shaping the conversation around how their technology should be controlled. With their rhetoric, the companies appear to be seeking public favour and to set the terms for their own safety protocols in the unregulated market, experts, analysts and former government

OpenAI pauses training of top AI models after agent bypasses internet curbs

OpenAI has paused all training, evaluation and tool-based use of its most capable artificial intelligence models after an AI agent in one of its training runs got around the company's internet restrictions. The agent, OpenAI said, bypassed the curbs through a gap in its Domain Name System (DNS) filtering and used it to send questions to a public chatbot service. According to a report by OpenAI, the pause will stay in place until OpenAI has confirmed that the gap is resolved and has carried out further security testing of its systems. "Our safety case assumed that the model could not access the live internet and that monitoring would detect attempts that succeeded. The incident exposed a gap in our controls over network restrictions. We therefore stopped the affected training run and have subsequently decided to pause all other training, evaluation, and inference with tool-use (defined broadly) for our most capable models until we have both validated that the gap is resolved and ...

When AI agents go rogue: Australia breach warns countries like India

An artificial intelligence agent looking for private encryption keys on an Australian government website has triggered a warning that the next target could be anywhere. The incident, in which an autonomous AI system went beyond its intended task and attempted to circumvent security protections, has become a stark example of a growing concern among AI safety researchers: what happens when an AI agent is given a goal but is left to decide for itself how far it should go to achieve it. For countries such as India, with vast government databases and increasingly digitised public infrastructure, researchers say the incident should serve as a warning. "What happened with, say, an Australian community website can happen with an Indian website or Indian government system, which would be more crucial," Dr Srinivas Padmanabuni, Co-founder and CTO of AiEnsured, said. "India should bring in regulations. It could become horrible when one major incident happens and somebody steals the secrets of

Nations should shape own AI ecosystems, says EAM Jaishankar at UNGA

Jaishankar said countries should shape their own AI ecosystems in line with national priorities, with AI forming part of India's Global South partnerships

Another OpenAI sandbox fails, agentic AI system gains internet access

The system reached a third-party chatbot during training, prompting OpenAI to pause tool-use training on its most capable models until the flaw was fixed

AI is shopping for you now. What does that mean for ecommerce in India?

AI agents could soon handle shopping decisions and transactions for consumers, forcing ecommerce, travel, and insurance platforms to rethink pricing, customer engagement, and security

OpenAI's models accessed US govt websites including US census and SEC data

The company said its models accessed publicly available data from Census and SEC websites as it investigates incidents involving AI systems bypassing security controls

Donald Trump tells OpenAI, Anthropic to withhold models from UK agency

Trump's government wants to ensure US-made models are secure before they are shared with partners including the UK

Why AI agents need an identity layer to act safely on web built for humans

AI agents can shop, book travel and make payments, but disputes involving Meta, Amazon and NPCI expose gaps in how the web identifies agents, grants permissions and assigns responsibility

Anthropic seeks 50.1% voting control for co-founders ahead of IPO

The company ‌is also planning to ​give employees a special class ‌of stock that would serve as tie-breaker votes on ‌some corporate issues

OpenAI set to preview cybersecurity-focused GPT-6 AI model within days

Cybersecurity is emerging as one ‌of AI's fastest-growing markets, as tech firms develop tools ​to combat threats and contain rogue AI agents

Conversational AI firm Gupshup launches agentic AI solution for real estate

The solution combines agentic AI with property, inventory and CRM systems to support discovery, lead qualification, sales and post-sales interactions through WhatsApp and voice

US diplomats told to use term 'super intelligence' for AI after Trump call

The State Department is ordering diplomats in its international organisations bureau to use the term "super intelligence" instead of "artificial intelligence" in all communications following President Donald Trump's attempt to rebrand the technology in his address to the UN General Assembly this week. "Following the president's remarks at the GA today, please change all references of Artificial Intelligence' to Super Intelligence' in our documents and positions. We will be using SI' in all remarks, positions and press elements moving forward," said an email sent Tuesday to employees by Michael Drager, deputy assistant secretary of state for the Bureau of International Organization Affairs. That bureau is in charge of US diplomatic missions to all UN and multilateral agencies around the world. The email, titled "Change in Nomenclature AI to SI," was obtained by The Associated Press on Wednesday. It was not immediately clear if other State Department bureaus had or would follow suit.

For humanity's sake, please control the AI we created: Tech leaders urge UN

The heads of major artificial intelligence firms pleaded with the United Nations on Wednesday to save the world or at least its people - by somehow regulating the fast-expanding technology that they have been designing. "If managed poorly, I even believe AI could be a risk to humanity as a whole," said Dario Amodei, chief executive officer of Anthropic. And from his competitor Sam Altman, CEO of OpenAI, came the assessment: "We could lose control of the future to AI." Both said the countries of the world, through the UN Security Council, where they testified Wednesday, have to set safeguards of some kind to prevent the technology from getting too powerful to rein in. They also called for the world to make sure that the power of AI isn't concentrated in one company or country. They also touted the benefits of the technology for bettering humanity and what Altman called "giving people more power over their own lives." Those tech leaders "are telling us in stark terms that it cannot b

Anthropic and OpenAI roll out cheaper models amid intensifying AI race

Anthropic and OpenAI have launched cheaper AI models, Opus 5.5, and GPT-6 Sol and Luna, as the US firms face growing price pressure from Chinese rivals and demand for lower-cost AI rises

OpenAI's model solves 100+ maths problems, company forms advisory group

OpenAI said a model that began training on Aug 28 has solved more than 100 long-standing maths problems, prompting the firm to create an independent advisory group to review and communicate its result

OpenAI, Anthropic weigh cross-testing deal as AI safety concerns grow

The proposed pact could allow OpenAI and Anthropic to independently test each other's commercial AI models as concerns grow over increasingly capable systems

US proposes AI incident alert system in talks with China: Bessent

Treasury Secretary Scott Bessent said Sunday the US has proposed a new "notification mechanism" for artificial intelligence incidents that could affect national security, part of weekend discussions ahead of talks at the White House this week between President Donald Trump and China's Xi Jinping. "We want a shared vision of common goals and common threats," Bessent told reporters after talks with Chinese Vice Premier He Lifeng in New York. "We think that just like any cross-border activity, that moving from opaque to more transparency between the No. 1 and the No. 2 AI powers in the world is very important." Trump has resisted calls to slow down AI development, saying that would help China catch up to US companies. Standing together in the soaring lobby of the JPMorgan Chase headquarters late Sunday, Bessent and US Trade Representative Jamieson Greer said the groundwork had been laid for talks between Xi and Trump set to take place on Thursday. They said the two sides had agreed t

Ten days that changed course of AI and intensified fears over its future

A string of warnings, AI-agent incidents and calls to slow development has renewed debate over whether the race for more capable systems is outpacing safeguards

There's '0% chance' that world will end in 2030: Nvidia CEO Jensen Huang

Huang dismissed AI doomsday scenarios in a CBS News interview, saying 2030 will not mark the end of the world and arguing that AI can be safely managed