GCHQ says AI cyber attacks by China and Russia will be routine by 2028

In short: The UK’s cyber security agency says AI will make state-backed cyber attacks from China and Russia a regular part of life by 2028.

What's going on

The National Cyber Security Centre (NCSC), part of GCHQ, says the UK is already facing cyber attacks that use AI. It expects these attacks to grow in speed and scale over the next two years.

In its annual review, the NCSC said it is “highly likely” that by 2028 China, Russia, and other countries will routinely use AI in cyber attacks to meet political and military goals. It said AI is making it easier for attackers to find weak spots in software, pick targets, and run attacks more efficiently.

The agency also expects to see AI-powered ransomware aimed at UK infrastructure within the next two years. Ransomware is a type of attack where criminals lock up an organisation’s computer systems and demand payment to unlock them, like putting a padlock on a business and charging for the key.

The NCSC said hackers are not usually letting AI run attacks on its own. Instead, they use it for “reconnaissance” (doing homework on a target), writing more convincing scam messages, and spotting security gaps.

The report said China is the most sophisticated and sustained threat to the UK, while Russia is the most audacious. It cited a December 2025 attempt on Poland’s energy grid that could have affected 500,000 people if it had worked.

What to watch

NCSC chief executive Richard Horne is expected to warn that UK defenses, especially for critical infrastructure like energy and transport, are not strong enough. He argues that organisations have relied on cheap tech for too long and should spend more on security and avoid connecting systems to the internet unless they need to be.

Source: Financial Times