On August 17, 2026, cybersecurity became one of the key issues for artificial intelligence users. Experts are recording a sharp increase in attacks on popular chatbot accounts, including ChatGPT, Claude, and Perplexity. Hackers are exploiting vulnerabilities in authentication systems to gain access to personal data, chat history, and even generated materials. In this article, we will examine in detail how to check for unauthorized devices in your profile and protect your data.

How to Check Active Sessions in ChatGPT

To check access to your ChatGPT (OpenAI) account, you need to perform a series of steps. In your computer browser, click on your username in the bottom left corner, then go to the "Settings" section. Next, select "Security and Login" and open the "Active Sessions" tab. Here you will see a list of all devices from which you have logged into your account. If you find an unfamiliar device, you can terminate it individually or click the "Log out of all devices" button to reset all sessions completely.

Protecting Claude and Perplexity Accounts

The Claude platform offers similar tools for access control. To check active sessions, open settings via your username in the bottom left corner (Settings -> Account) and view the "Active sessions" section. To log out of a specific device, hover over the session, click the three dots on the right, and select "Log out" or "Terminate". The "Log out of all devices" option is also available. Since Claude operates without passwords, re-entry is done via a link sent to your email.

The Perplexity service does not display a list of specific devices but provides similar protection mechanisms. If you suspect a hack, go to "All settings" via the user menu, click "Sign out of all sessions", and confirm the action. Re-entry is performed using a one-time six-digit code or a direct link via email.

Contradictory Data

There are disagreements among experts regarding the effectiveness of current security measures. Some specialists claim that two-factor authentication (2FA) is sufficient to protect accounts, while others point to the need to implement more complex methods, such as biometric verification. Furthermore, there are contradictions regarding the dates of the introduction of new security features: some sources report the launch of updates in July 2026, while others say August.

Data Protection Recommendations

To maximize the protection of your accounts, it is recommended to regularly check active sessions, use complex passwords, and enable two-factor authentication. It is also important to be cautious when clicking links in emails and messages, as phishing attacks remain one of the main methods for gaining access to accounts. If you suspect a hack, change your password immediately and notify the platform's support service.

Conclusion

In the context of the growing threat of cyberattacks on AI platforms, users need to be particularly attentive to the security of their accounts. Regularly checking active sessions, using strong passwords, and two-factor authentication will help minimize the risks of hacking. Remember that data protection is not only the responsibility of developers but also of every user.