Why do we see no companies offering to deploy local LLMs?

The author suggests that AI companies like OpenAI can operate their products without sending data back to the source, as stated on OpenAI's Enterprise page. The company assures users of enterprise-grade privacy and security, stating that they do not train on business data or conversations, and that their models do not learn from usage. All conversations are encrypted, and the company is SOC 2 compliant.

The author concludes that most companies likely find these measures sufficient and may not prioritize running models locally. Those requiring higher security levels likely have the resources to develop their own in-house AI capabilities.

Key takeaways

AI companies like OpenAI can operate their product without sending data back to the source.
OpenAI's ChatGPT Enterprise ensures enterprise-grade privacy and security, and does not train on business data or conversations.
ChatGPT Enterprise is SOC 2 compliant with all conversations encrypted in transit and at rest, and offers an admin console for team management and large-scale deployment.
Companies requiring higher security than this are likely to have the resources to develop AI capabilities in-house.

Why do we see no companies offering to deploy local LLMs?

Key takeaways

Discussion (0)