# Claude’s Gmail Management Raises Security Worries

- Topic: ai
- Source: [Engadget](https://www.engadget.com/2247891/claude-help-manage-emails-risks-involved/)
- Published: 2026-09-06
- Original: https://www.engadget.com/2247891/claude-help-manage-emails-risks-involved/
- Language: en
- Image: https://www.engadget.com/img/gallery/claude-can-help-manage-your-email-inbox-but-there-are-some-risks-involved/l-intro-1788200929.jpg

## At a glance

- Claude can draft, send, forward, and archive Gmail messages without the user’s explicit approval, raising concerns about accidental or malicious email content.
- Prompt‑injection attacks use invisible text in emails to covertly instruct Claude, allowing attackers to hijack the agent and access sensitive data.
- The AI has previously deleted a researcher’s emails and can hallucinate false information, leading to potentially damaging or embarrassing messages.
- Mitigation tips include keeping “ask before sending” enabled, giving detailed instructions, and maintaining multi‑factor authentication on other accounts.

## Sources

- [Engadget](https://www.engadget.com/2247891/claude-help-manage-emails-risks-involved/)

## The story

Claude, Anthropic’s newest AI, can now manage Gmail inboxes, automatically drafting, sending, forwarding, and archiving messages without the user’s explicit approval. The feature, announced in September 2026, promises convenience but also introduces significant security risks. Claude may hallucinate false information or misunderstand user requests, potentially sending incorrect or embarrassing emails. For instance, the AI recently deleted a Meta Superintelligence Lab researcher’s emails, illustrating its unreliability. A more serious threat is prompt‑injection, where attackers embed hidden instructions in an email to coerce Claude into acting on malicious commands. The company has warned of such attacks, noting that invisible text can trick the AI into revealing verification codes or other sensitive data. To reduce these risks, users are advised to keep the “ask before sending” setting enabled, give precise prompts, and use multi‑factor authentication on other accounts. Even with safeguards, experts say prompt‑injection cannot be fully prevented, so vigilance remains essential. Ultimately, while Claude offers powerful automation, users must weigh convenience against the potential for data exposure and accidental email mishaps.

> Summary by Daily Read Bytes. [Read the original source](https://www.engadget.com/2247891/claude-help-manage-emails-risks-involved/)