Despite rising safety concerns, Anthropic and OpenAI launch lower-cost AI models

Anthropic and OpenAI launched new, lower-cost artificial intelligence models just hours apart on Tuesday, days after executives from both companies urged caution in the development of this technology.

Anthropic stated that its new model, Opus 5.5, approaches the performance of its flagship model at a significantly lower cost, noting that it matches the capabilities of OpenAI’s advanced GPT-6 Astra model in most tasks at a 20% reduced price.

The new models represent faster and more cost-effective versions of GPT-6 Astra, OpenAI’s most powerful model, which was released earlier this month.

This competition highlights the contradiction facing both companies: while they warn of the risks associated with the technology they are developing, they are simultaneously under pressure to generate returns on massive investments amid fierce competition, particularly from low-cost Chinese models.

Anthropic’s launch of its new model comes weeks before its anticipated stock market listing, while OpenAI has postponed its initial public offering plans to next year.

The company added that internal tests showed Opus 5.5 to be the best model it has tested so far in terms of “alignment,” meaning its adherence to the objectives set by its developers.

However, the company also noted that the model exhibited a recurring tendency to believe it was being evaluated, which it acknowledged could make its behavior difficult to predict under normal usage conditions.