Palantir, NVIDIA and Microsoft shift AI partnerships over data security
Palantir and NVIDIA are rethinking how they work with big AI players like Anthropic and OpenAI, all because of concerns about keeping their data safe.
Palantir has pressed Anthropic for irrevocable zero-data-retention guarantees, while NVIDIA limits Anthropic's use to less sensitive tasks and relies on its own Nemotron models for internal work.
Microsoft, meanwhile, is stepping up by offering more secure cloud options and its own AI tools.
Palantir wants 0 retention from Anthropic
Palantir wants zero data retention from Anthropic before using its models. NVIDIA is only letting Anthropic handle less sensitive tasks and prefers using its own Nemotron models for internal work.
These moves come after criticism of both Anthropic and OpenAI for how they handle user data, though both do not train on customer data by default unless companies opt in, though they collect anonymised metadata to improve products.
With all this in mind, Anthropic CEO Dario Amodei has even called for slowing down AI development (a sentiment echoed by leaders like Elon Musk) out of concern for where unchecked AI progress might lead.