OpenAI releases GPT-6 Astra and says it surpasses Anthropic

In short: OpenAI has launched a new AI model called GPT-6 Astra and says it is now ahead of rival Anthropic.

What happened

OpenAI, the company behind ChatGPT, announced a new AI model on Thursday called GPT-6 Astra. OpenAI said Astra is “market-leading” in software engineering, science, and cyber security.

The company is trying to show it has regained momentum against Anthropic, a close competitor started by former OpenAI staff. According to the Financial Times report, Anthropic has recently told investors it has been leading this year, and it has reached a higher valuation ahead of a possible stock market listing.

OpenAI president Greg Brockman also said Astra could be considered “artificial general intelligence,” or AGI. AGI is a fuzzy term, but it usually means AI that can outperform humans at many different thinking tasks, instead of being good at only one thing (like a Swiss Army knife rather than a single tool).

OpenAI said Astra will initially be given to a small group of businesses. The goal is to give those companies time to address cyber security concerns before the model becomes more widely available “over the coming days.” This matters because more capable AI systems, especially “AI agents” (tools that can take actions with little human input), can also be misused or behave in risky ways.

Why it matters

For regular people, this competition can affect what AI tools can do at work and how much they cost. OpenAI says Astra is more efficient, meaning the “price per task” could be lower, and it highlighted uses like financial modelling, tax preparation, data analysis, and filling in forms.

Source: Financial Times