AI and network security

Can AI hack my website?

Yes. AI systems can now scan websites, find common flaws such as SQL injection, cross-site scripting and exposed secrets, and in some cases exploit them with little human help. They mostly find the same kinds of weakness human attackers do, but faster and across far more sites, so an unpatched or overexposed website is found sooner than it used to be.

TL;DR

AI can hack websites with the usual weaknesses, faster and at greater scale than people. An autonomous AI tester topped HackerOne's US ranking in 2025, and the average time to exploit fell to five days in 2023. Patch fast, remove what need not be public and use phishing-resistant admin sign-in. Operational devices can stay on local networks, out of reach.

Learning objectives

After reading this article you will be able to:

  • Describe what AI systems have been shown to find and exploit in real software and websites
  • Protect a website by patching quickly, shrinking what it exposes and keeping secrets out of reach
  • Separate the public website from the operational systems that do not need to be on the internet

AI website hacking examples

AI systems are now finding and exploiting real flaws in real software, and attackers have used them in live campaigns.

  • On bug bounty platforms. In June 2025, XBOW reported that its fully autonomous penetration tester had climbed to the top of HackerOne’s US ranking, competing alongside thousands of human researchers and making nearly 1,060 vulnerability submissions. The findings included remote code execution, SQL injection, XML external entities, path traversal, server-side request forgery, cross-site scripting, cache poisoning and exposed secrets.
  • In widely used code. In October 2024, Google Project Zero’s Big Sleep agent found an exploitable memory-safety flaw in SQLite before it appeared in an official release, which the team called the first public example of an AI agent finding such a flaw in widely used real-world software.
  • In a competition on real projects. At DARPA’s AI Cyber Challenge final in August 2025, competing systems found 54 of 63 planted vulnerabilities, patched 43, and also discovered 18 real, previously unknown ones.
  • In an attack. Anthropic reported in November 2025 that a state-sponsored group used its model to perform 80 to 90 percent of an espionage campaign against about thirty targets, with humans stepping in at only a handful of decision points. The attackers got past safeguards by splitting the work into small tasks that looked harmless on their own.

AI attack speed and scale

AI changes the speed and scale of website attacks more than the kinds of attack. AI tools mostly find the weaknesses human attackers always looked for. What changes is how many sites they can test and how quickly they act on a newly published flaw.

That speed was already rising before these tools spread. Google’s threat intelligence analysis found that the average time between a vulnerability’s disclosure and its exploitation fell from 32 days in 2021 and 2022 to five days in 2023. A site that patches monthly now spends most of each month exposed to anything published since the last update.

Website protection against AI attacks

Protecting a website from AI-driven attacks means closing the common doors, and closing them faster.

  1. Patch internet-facing software within days. Content management systems, plugins, frameworks and web servers first. CISA’s Known Exploited Vulnerabilities catalog lists flaws already being used in attacks; anything on it that you run comes first.
  2. Remove what does not need to be public. Staging sites, old admin panels, test APIs, debug endpoints and unused plugins are all attack surface. Put admin interfaces behind your identity provider or a VPN instead of the open internet.
  3. Use phishing-resistant sign-in for admins. CISA calls phishing-resistant MFA, such as FIDO and WebAuthn security keys or passkeys, the gold standard. Stolen passwords are cheap, and AI makes phishing pages more convincing.
  4. Keep secrets out of the front end and out of code. Any key in a browser bundle can be read. Store server keys in a secret manager and scope each to one job.
  5. Fix the usual flaw classes in your own code. Validate input, use parameterised queries, encode output and check authorisation on every request, since these are exactly what automated testing finds.
  6. Test yourself first. Run automated scanning, or an AI-assisted test, against your own site before someone else does.

Operational systems off the internet

Most of the systems that run a business do not need to be on the internet at all. The website has to be public, but the devices and tools behind your operations, such as handhelds, scanners, point-of-sale terminals, sensors, kiosks and the systems that coordinate them, rarely need to be reachable from the internet. Each one that is reachable is one more thing an automated attacker can find.

Offline Protocol is built to take that layer off the internet. Devices discover each other, verify identities with Ed25519 keys they generate themselves, message over MLS end-to-end encryption, keep shared state in sync as replicated documents and call each other’s services over Bluetooth LE, Wi-Fi and local IP, with no internet and no API key. Your backend stays the system of record and receives results when it is reachable. An AI scanning the internet can find your website; it cannot scan devices that coordinate locally and have no public address to find.

Frequently asked questions

Can ChatGPT or Claude hack a website?

AI assistants have safeguards against clearly malicious requests, but attackers have got around them by splitting an attack into small, innocent-looking tasks. Anthropic reported in November 2025 that a state-sponsored group used its model this way to run 80 to 90 percent of an intrusion campaign against about thirty targets.

What do AI hacking tools usually find?

Mostly the familiar classes of web flaw. XBOW's reports to HackerOne included remote code execution, SQL injection, XML external entities, path traversal, server-side request forgery, cross-site scripting, cache poisoning and exposed secrets.

Is a small business website a target?

Yes. Automated tools do not choose targets by size; they scan for whatever is vulnerable. A small site running an outdated plugin is as easy to find as a large one.

Can AI also defend my website?

Yes. The same techniques find flaws for defenders. Google's Big Sleep agent found an exploitable SQLite flaw before it reached a release, and DARPA's AIxCC teams found and patched real vulnerabilities in open-source code.

How fast do I need to patch?

For internet-facing software, within days, not weeks. Google's analysis put the average time-to-exploit at five days in 2023, and CISA's Known Exploited Vulnerabilities catalog lists the flaws attackers are already using.

Sources

Build it with Offline Protocol

This page sets out which parts of a workflow the Offline Protocol SDK handles directly between devices and which stay with your backend, so you can see what can come off the public internet.

Read What the SDK handles