The world is now aware of the potential risks of AI models. These systems could be misused easily as they identify hidden critical software flaws and have the potential to hack systems on their own. Highlighting the risks of top frontier AI models, Lee Klarich, Chief Product and Technology Officer at Palo Alto Networks, warned that the top AI models from Anthropic and OpenAI are better at finding security flaws than the Palo Alto release. In simple words, he explains how the next major cyberattack could be triggered by top AI models.
Klarich explained how Palo Alto Networks started testingAnthropic’s Claude Mythoson April 7, 2026, as part of Project Glasswing. The company further tested AI systems like Claude Opus 4.7 and OpenAI’s GPT-5.5 Cyber under its Trusted Access for Cyber program.
The company revealed that it tested the AI tools and scanned more than 130 products across its platforms. It found that it had to release fixes for flaws covering nearly 75 security issues in its latest monthly security update.
Klarich warns that companies have only three to five months to improve their security before hackers begin to use these advanced tools. He highlighted that AI-powered cyberattacks could become common very soon. He explainedAI modelsspot different flaws, which is why using multiple AI systems together becomes necessary.
Klarich advised that businesses should begin using AI tools to scan their software and fix critical security bugs before cyberattackers find them. Moreover, he suggested that companies must reduce their online exposure, secure customer-facing systems and enhance protection across supply chains and cloud systems.
He recommended using powerful security tools like AI-powered threat detection, zero-trust systems and advanced endpoint protection to stop attacks faster. What lies ahead is the solution to the problem that emerged from AI could potentially be AI. Klarich said, “This is the light at the end of the tunnel: A future where software is secure by design.”
Amidst these concerns, Union Finance MinisterNirmala Sitharamanasked the Indian banks to strengthen their cyber defences against the potential risks associated with Mythos-like AI systems. She chaired a high-level meeting last month with banking officials and the Union Minister For Electronics and Information Technology Ashwini Vaishnaw along with key stakeholders to assess the impact of emerging artificial intelligence models and how such technologies can be misused to exploit software vulnerabilities.
Notably, she instructed the Indian Bank’s Association to prepare a coordinated institutional mechanism to ensure effective responses to cyber threats. She also asked the banking officials to engage top cybersecurity professionals and specialised agencies to make the monitoring and defence systems better.
What lies ahead is how it will be noteworthy how the banking sector, industries and government across the world will need measures to combat such increasing cybersecurity issues that arose from growing AI adoption that was once claimed to help humans.
GetLatest Newslive on Times Now along with Breaking News and Top Headlines fromTechnology Scienceand around the world.
Source: India Latest News, Breaking News Today, Top News Headlines | Times Now
