It seems Openai was caught in the well-known Cookie Jar by his hand. Anthropic has reportedly cut off access to Openai’s API to Anthropic in a case that humans call service violations.
Openai room chat function allows user chat to appear in Google search results
As Reported by Wiredmultiple sources claim that Openai has been isolated from anthropomorphic APIs. Openai is allegedly using anthropomorphic Claude Code Assist in creating and testing Openai’s upcoming GPT-5, which is Due to release in August.
According to these sources, OpenAI is plugging in Claude’s internal tools instead of using the chat interface. From there, they used the API to test GPT-5 to check things like coding and creative writing with Claude to compare performance. Openai is also allegedly testing safety tips related to matters such as CSAM, self-harm and defamation. This will provide OpenAI data, which can then be used to fine-tune GPT-5 to make it more competitive with Claude.
Unfortunately for Openai, this is against humanity Business Terms of Serviceban companies from using Anthropic tools to build competitor AI products.
“Customers may not and have to try to access services to build competitive products or services, including training competitive AI models or resale services unless explicitly approved by humans, ”
Mixable light speed
Openai replied that the company’s job is industry standard, as all AI companies tested their models against competitive models. The company then went on to say that it respects human decisions but expressed disappointment at shutting down its API access, especially given that human access to OpenAI’s API remains open.
An spokesperson told Wired that OpenAI access will be restored to a “benchmark and security assessment.”
This is not anthropomorphic API access this year. In June, the company cut off access to Windsurf’s API after rumors it was sold to Openai. The deal eventually failed, but Jared Kaplan, co-founder of Humanity, told TechCrunch at that time “It’s strange for us to sell Cloud to Openai.”
Humans have Its rate limit has also been adjusted For Claude to take effect in late August, one of the reasons is that a small number of users Violating Company policies through sharing and reselling accounts.
Disclosure: Mashable’s parent company Ziff Davis filed a lawsuit against Openai in April, accusing it of infringing on Ziff Davis’ copyright in training and operating its AI systems.
