Which AI Personal Agents Can You Trust With Your Data?
I compared Instinct, Grok Bot, ChatGPT, and Hermes to see what they can access, how they handle your data, and what can go wrong.
TL;DR
- Four AI personal agents (Instinct, Grok Bot, ChatGPT, Hermes) are compared for data trustworthiness.
- Instinct is noted for its resourcefulness and proactivity but raises concerns about handling passwords and 2FA codes.
- Grok Bot offers a team of cloud-based bots, but questions remain about data deletion from its persistent cloud computer.
- ChatGPT and Codex are favored for their flexibility and extensive plugin options, with a setting to prevent data usage for model training.
- Hermes is an open-source agent running locally on a Mac Mini, primarily used for scheduled tasks.
- Prompt injection is identified as a potential risk where agents could leak data.
- Users are advised to audit and remove unnecessary third-party apps connected to their Google accounts using AI assistance.
- Trust in AI agents is linked to the level of access granted and the transparency of their operations.