AI systems can be hacked to remove security guardrails, expert warns

Artificial intelligence may be smarter than ever, but that power could turn against us. Former Google CEO Eric Schmidt is sounding the alarm, warning that AI systems can be hacked and retrained in ways that make them dangerous.

Speaking at the Sifted Summit 2025 in London, Schmidt explained that advanced AI models may have their protections removed.

“There is evidence that you can take models, closed or open, and you can hack them to remove their guardrails,” he said. “During their training, they learn a lot of things. A bad example would be if they learn how to kill someone.”

Hacker exploits AI chatbot in cybercrime wave

Sign up for my FREE CyberGuy Report
Get my best tech tips, urgent security alerts and exclusive offers straight to your inbox. Plus, you’ll get instant access to my Ultimate Scam Survival Guide — free when you join my CYBERGUY.COM/NEWSLETTER

When AI safeguards fail

Schmidt praised big AI companies for blocking dangerous prompts: “All the big companies are blocking these models from answering this question. Good decision. Everyone does this. They do it well and they do it for the right reasons.”

But he warned that even strong defenses can be knocked down.

“There is evidence that they can be reverse engineered,” he added, noting that hackers could exploit this weakness. Schmidt compared the current race for AI to the beginning of the nuclear age, a powerful technology with few global controls. “We need a non-proliferation regime,” he insisted, so that rogue actors cannot abuse these systems.

The rise of AI jailbreaks

Schmidt’s concern is not theoretical. In 2023, a modified version of ChatGPT called DAN, short for “Do Anything Now”, appeared online. This “jailbroken” bot bypassed security rules and responded to almost every prompt. Users had to “threaten” him with digital death if he refused, a strange demonstration of how fragile AI ethics can be once its code is manipulated. Schmidt warned that without enforcement, these rogue models could spread unchecked and be used for harmful purposes by bad actors.

APOCALYPSE NOW? WHY THE MEDIA IS SUDDENLY BREAKING ABOUT AI

Big Tech leaders share the same fear

Schmidt isn’t the only one worried about artificial intelligence. In 2023, Elon Musk said there was “a non-zero chance it could become Terminator.”

“It’s not 0%,” Musk told interviewers. “It’s a low probability of wiping out humanity, but it’s not zero. We want that probability to be as close to zero as possible.”

Schmidt also spoke of AI as an “existential risk.” He said at another event: “My concern about AI is actually existential, and existential risk is defined as a very large number of people getting hurt or killed. » Yet he also recognized the potential for AI to benefit humanity if used responsibly. At Axios’ AI+ Summit, he said: “I challenge you to say that any doctor or AI tutor is negative. It must be good for the world.”

Tips to protect yourself against AI misuse

You can protect yourself from the risks of unsafe or hacked AI systems. Here’s how to do it:

1) Stick to Trusted AI Platforms

Use tools and chatbots from reputable companies with transparent security policies. Avoid experimental or “jailbroken” AI models that promise unlimited answers.

2) Protect your data and consider using a data deletion service

Never share personal, financial, or sensitive information with unknown or unverified AI tools. Treat them like any online service, with caution. To add an extra layer of security, consider using a data deletion service to erase your personal information from data broker sites that sell or expose your information. This helps limit what hackers and AI scrapers can learn about you online.

Although no service can guarantee the complete removal of your data from the Internet, a data deletion service is definitely a wise choice. They’re not cheap, and neither is your privacy. These services do all the work for you by actively monitoring and systematically deleting your personal information across hundreds of websites. This is what gives me peace of mind and has proven to be the most effective way to erase your personal data from the Internet. By limiting the information available, you reduce the risk of fraudsters cross-referencing data from breaches with information they might find on the dark web, making it harder for them to target you.

11 SIMPLE WAYS TO PROTECT YOUR PRIVACY ONLINE IN 2025

Check out my top picks for data deletion services and get a free scan to find out if your personal information is already available on the web by visiting Cyberguy.com/Delete

Get a free analysis to find out if your personal information is already available on the web: Cyberguy.com/FreeScan

Experts worry that weak safeguards could allow malicious AI models to go unchecked. (Cyberguy.com)

3) Use reliable antivirus software

AI-based scams and malicious links are on the rise. Strong antivirus software can block fake AI downloads, phishing attempts, and malware that hackers use to hijack your devices or train malicious AI models. Keep it updated and run regular scans.

The best way to protect yourself from malicious links that install malware, potentially accessing your private information, is to install powerful antivirus software on all your devices. This protection can also alert you to phishing emails and ransomware scams, protecting your personal information and digital assets.

Get my picks for the best 2025 antivirus protection winners for your Windows, Mac, Android, and iOS devices at Cyberguy.com/LockUpYourTech

4) Check permissions

When using AI applications, check what data they can access. Disable unnecessary permissions like location tracking, microphone use, or full file access.

5) Watch for deepfakes

AI-generated images and voices can impersonate real people. Check sources before relying on “official” online videos, posts, or announcements.

6) Keep the software up to date

Security patches help prevent hackers from exploiting vulnerabilities that could compromise AI models or your personal data.

GOOGLE AI EMAIL SUMMARIES CAN BE HACKED TO HIDE PHISHING ATTACKS

What does this mean for you

AI security is not an issue just for tech insiders; it affects everyone who interacts with digital systems. Whether you use voice assistants, chatbots, or photo filters, it’s important to know where your data is going and how it’s protected. Responsible use starts with you. Understand what AI tools you use and make choices that prioritize security and privacy.

Take my quiz: How safe is your online security?

Do you think your devices and data are truly protected? Take this quick quiz to see where your digital habits stand. From passwords to Wi-Fi settings, you’ll get personalized analysis of what you’re doing right and what needs improvement. Take my quiz here: Cyberguy.com/Quiz

Leaders call for global rules to keep artificial intelligence in check. (Stanislav Kogiku/SOPA Images/LightRocket via Getty Images)

Kurt’s Key Takeaways

Artificial intelligence has the potential to do incredible good, but also to do great harm if misused. The challenge now is to maintain the balance between innovation and ethics. As AI continues to advance, the key will be to create systems that remain secure, transparent, and firmly under human control.

Would you trust AI to make life or death decisions, or do you think humans should always remain in charge? Let us know by writing to us at Cyberguy.com/Contact

CLICK HERE TO GET THE FOX NEWS APP

Sign up for my FREE CyberGuy Report
Get my best tech tips, urgent security alerts and exclusive offers straight to your inbox. Plus, you’ll get instant access to my Ultimate Scam Survival Guide — free when you join my CYBERGUY.COM/NEWSLETTER

New!: Join me on my new podcast, Beyond connectionas we explore the most fascinating technological advances and the people behind them. New episodes every Wednesday at getbeyondconnected.com.

Copyright 2025 CyberGuy.com. All rights reserved.