Synopsis
Cybersecurity faces a monumental opportunity and challenge with advanced AI models now identifying software flaws at lightning speed. Zscaler CEO Jay Chaudhry highlights that vulnerabilities hidden for decades are being uncovered, demanding rapid fixes before exploitation. The company is actively testing AI, including Anthropic’s Mythos, to stay ahead in this rapidly evolving threat landscape, emphasizing the urgent need for enhanced cybersecurity measures.
The rapid rise of advanced artificial intelligence models capable of identifying software vulnerabilities at unprecedented speed is creating one of the biggest opportunities the cybersecurity industry has seen in years, according to Zscaler founder and chief executive Jay Chaudhry.“There has never been a time in cyber where there’s more need for cybersecurity than we see today,” Chaudhry told ET in an interview.The emergence of these models is fundamentally changing the cybersecurity landscape because vulnerabilities that once took researchers weeks or months to uncover can now be identified far more quickly, giving defenders and attackers access to the same powerful capabilities.“What we’ve learned over the past few months is that these models can find vulnerabilities in software at a much faster speed, and some of the vulnerabilities have been hidden for 10 or 20 years,” Chaudhry said during the interview on the sidelines of Zscaler’s Zenith Live cybersecurity conference. “That creates a new challenge. How do we fix those vulnerabilities before they get exploited?”The San Jose, California-headquartered company has a market capitalisation of roughly $20 billion.Zscaler was among a group of 40 companies that received access to Anthropic’s advanced AI model, Mythos, under a controlled testing programme. The company used the model to scan its own software and identify vulnerabilities before they could be exploited.“We’ve been running the model and comparing it against earlier versions as well as OpenAI’s latest models,” Chaudhry said. “The preview model is a few steps ahead, but it’s a matter of time before other models catch up.”Yet, he believes concerns around restricted access to advanced AI models are overblown.“The focus on who gets access is probably misplaced,” Chaudhry argued. “Other models can already find 80% of what it does. Customers need to focus on fixing what today’s models can already identify rather than waiting for access to one specific model.”He also criticised recent US restrictions around advanced AI systems.“I personally think that was a knee-jerk reaction,” he said. “Ideally, these issues should have been sorted out before the models were made available.”The restrictions have intensified discussions around sovereign AI, with countries increasingly looking to reduce dependence on foreign technology providers.“Anything that becomes mission critical for a country requires sovereignty,” Chaudhry said. “First it was sovereign cloud. Sovereign AI is the natural extension of that.”India, he believes, is still behind the US and China in AI but is making meaningful progress.“The effort being put in place by the Prime Minister and the IT ministry is pretty impressive,” he said. “Twenty years ago, you could not imagine the government taking proactive steps to make GPUs available or attract semiconductor investments.”He also sees changing US immigration policies potentially benefiting India.“India has the best talent,” Chaudhry said. “With the latest US policies, some of the brain drain could reverse, and that can help India catch up faster in AI.”India already plays a central role in Zscaler’s operations, with nearly 40% of its workforce based in the country.Chaudhry, who comes from humble roots in a small village in India, has built five successful technology companies and is widely regarded as the richest Indian-American in the US. He credits his success to values instilled by his parents.“I learned three things from my parents that I never forgot—hard work was number one, honesty was number two and integrity was number three,” he said.Zero-Trust Architecture“We have come up with two thoughts. One, how can I prevent breaches without having to fix every vulnerability immediately? Two, if a breach happens, how do I minimise the impact of that breach,” Chaudhry said.According to him, organisations need to move away from legacy security tools built around firewalls and virtual private networks (VPNs).“Anything that can be seen from the internet can be discovered by bad actors,” he noted. “In the zero-trust world, those firewalls and VPNs are no longer needed. If they can’t find you, they can’t attack you.”The growing concern around AI-powered cyber threats is translating into stronger customer engagement, although Chaudhry cautioned that the commercial impact will take time to materialise.“There is urgency and worry at the board level,” he said. “Boards are asking CISOs and CIOs what is going on and what needs to be done. But this is not a situation like Covid that needs to be addressed tomorrow.”The reporter was in Vienna at the invitation of Zscaler(Tanya Pandey contributed to this story)
(Catch all the Technology News News, and Latest News Updates on The Economic Times.)
…more
Elevate your knowledge and leadership skills at a cost cheaper than your daily tea.
