Things will 'get a lot worse': Anthropic security chief issues stark warning on digital security
Until these advanced security tools are widely used in everyday software development, attackers can exploit existing AI capabilities to discover weaknesses in older software, Jason D. Clinton said.
- Oct 23, 2026,
- Updated Oct 23, 2026 11:17 AM IST
In the last couple of days, the world has been witnessing major cyberattacks and vulnerability issues, raising concerns about the growing sophistication of digital threats. Amid concerns, Jason D. Clinton, Anthropic’s Deputy Chief Information Security Officer (CISO), shared a stark warning that things will get “a lot worse before they get better.”
He said the tech industry is in a difficult period, as companies may find it increasingly hard to protect their systems, customer data, and online services. In an X post, Clinton said, “As bad as things have gotten in the past 4 days, they are going to get a lot worse before they get better,” referring to the recent scares of severe digital incidents.
Must read: BT BIG STORY: What happens when the keys to internet security are no longer secret
“Security leaders, never let a crisis go to waste: please go to your boards and ask for help right now. Not for AI, for the basics.”
He wrote the post, highlighting his post from August, in which he said: “The vulnerability tsunami is coming in just a few months, and there won’t be enough time to patch critical infrastructure.” The massive wave of cyber threats that Clinton warned about has now arrived. Now, it's time for companies and security leaders to push for stronger cybersecurity measures.
He further urged the tech industry to “hold on for ~18 months.” Furthermore, in a reply to an X user, Clinton highlighted a term “vulnzero,” which is used for zero vulnerabilities. He predicts that as AI capabilities are rapidly expanding, advanced AI security tools will soon become cheaper. With such tools, developers could identify vulnerabilities and fix them before the code goes live in production. This could make software more secure from the outset.
Must read: eMudhra, PHP security scare: Govt to hold high-level review, audit to assess potential damage
However, until these advanced security tools are widely used in everyday software development, attackers can exploit existing AI capabilities to discover weaknesses in older software, automate the process of finding security bugs, exploit known vulnerabilities faster than human security teams, and exploit systems before developers have time to release patches.
The forecast sparked a wave of criticism, with many arguing that AI companies were making cyber threats worse before defences are in place.
For Unparalleled coverage of India's Businesses and Economy – Subscribe to Business Today Magazine
In the last couple of days, the world has been witnessing major cyberattacks and vulnerability issues, raising concerns about the growing sophistication of digital threats. Amid concerns, Jason D. Clinton, Anthropic’s Deputy Chief Information Security Officer (CISO), shared a stark warning that things will get “a lot worse before they get better.”
He said the tech industry is in a difficult period, as companies may find it increasingly hard to protect their systems, customer data, and online services. In an X post, Clinton said, “As bad as things have gotten in the past 4 days, they are going to get a lot worse before they get better,” referring to the recent scares of severe digital incidents.
Must read: BT BIG STORY: What happens when the keys to internet security are no longer secret
“Security leaders, never let a crisis go to waste: please go to your boards and ask for help right now. Not for AI, for the basics.”
He wrote the post, highlighting his post from August, in which he said: “The vulnerability tsunami is coming in just a few months, and there won’t be enough time to patch critical infrastructure.” The massive wave of cyber threats that Clinton warned about has now arrived. Now, it's time for companies and security leaders to push for stronger cybersecurity measures.
He further urged the tech industry to “hold on for ~18 months.” Furthermore, in a reply to an X user, Clinton highlighted a term “vulnzero,” which is used for zero vulnerabilities. He predicts that as AI capabilities are rapidly expanding, advanced AI security tools will soon become cheaper. With such tools, developers could identify vulnerabilities and fix them before the code goes live in production. This could make software more secure from the outset.
Must read: eMudhra, PHP security scare: Govt to hold high-level review, audit to assess potential damage
However, until these advanced security tools are widely used in everyday software development, attackers can exploit existing AI capabilities to discover weaknesses in older software, automate the process of finding security bugs, exploit known vulnerabilities faster than human security teams, and exploit systems before developers have time to release patches.
The forecast sparked a wave of criticism, with many arguing that AI companies were making cyber threats worse before defences are in place.
For Unparalleled coverage of India's Businesses and Economy – Subscribe to Business Today Magazine