Advertisement

GPT-6 Astra: OpenAI introduces its most advanced AI model

Stay updated - Follow TOK on WhatsApp for instant alerts!
OpenAI has launched GPT-6 Astra, its most capable AI model yet, with major advances in reasoning, coding, research and computer use.
The model achieved a 100% score on ExploitBench and reached the Critical cybersecurity capability level under OpenAI's safety framework.
Astra is initially rolling out to selected organizations before expanding to ChatGPT paid plans and the OpenAI API.
0:00 / --:--
Advertisement

OpenAI has introduced GPT-6 Astra, its most powerful and advanced flagship artificial intelligence (AI) model, describing the system as a major step towards the development of artificial general intelligence (AGI).

Follow Times of Karachi on Google News and add as a preferred Source on Google

AGI refers to AI systems capable of learning and performing a wide range of tasks that currently require human intelligence.

According to OpenAI, GPT-6 Astra has been developed to assist with software development, research, autonomous agent workflows and advanced reasoning.

The company said the new model is also capable of identifying previously hidden security vulnerabilities, adding that it has been designed to handle increasingly complex tasks with greater autonomy.

GPT-6 Astra can process both written instructions and image-based prompts and is capable of producing structured responses and working with streaming outputs, according to the company.

Advertisement

OpenAI said internal testing showed GPT-6 Astra achieving a 100% score on ExploitBench, demonstrating its ability to identify and address sophisticated cybersecurity challenges.

Alongside the model’s enhanced capabilities, OpenAI said it has introduced additional security measures to prevent misuse and reduce potential risks associated with the system.

The company also claimed that GPT-6 Astra is significantly harder to jailbreak than GPT-5.6, strengthening protections against attempts to bypass its safety safeguards.

The launch represents another step in OpenAI’s efforts to develop increasingly capable AI systems that can reason, conduct research, write software and perform complex tasks with greater independence.

Earlier to this, Claude, the artificial intelligence (AI) chatbot developed by Anthropic, experienced a major outage that left users unable to access its services, causing disruption for businesses, software developers, and other customers who rely on the platform.

Advertisement
Share

Follow Times of Karachi on Google News and explore your favorite content more quickly!
Leave a Reply
Related Posts
🚫 Ad blocker detected. Please disable your ad blocker to support our content.
Close Button
Advertisement