Chinese and Russian AI cyber attacks will be routine by 2028, says GCHQ

The UK's National Cyber Security Centre, part of GCHQ, says in its annual review published on Tuesday that 'by 2028, it is highly likely China, Russia, and other nation states will routinely use AI cyber to achieve their geopolitical goals', with the speed and scale of attacks rising over the next two years and AI-powered ransomware attacks on UK infrastructure expected within two years. So far hackers use AI mainly for reconnaissance, tailored phishing and finding and exploiting vulnerabilities, not to fully automate attacks. The NCSC received 1,831 reports of attacks this year, 80 of them nationally significant. It calls China 'most sophisticated and sustained' and Russia 'most audacious', citing the December 2025 attack on Poland's energy grid, attributed to a Russian security services cyber unit, which would have affected 500,000 people. The FT places this alongside Monday's hack of supermarket freezers across Europe and last week's $6.3bn Binance theft. NCSC chief executive Richard Horne is expected to say 'for decades we have collectively got away with tech on the cheap' and that the danger is 'not because AI has invented new ways to attack us, but because vulnerabilities that have been tolerated for far too long can now be discovered and exploited at a scale that was once unimaginable', urging more investment in critical-infrastructure defences and fewer unnecessary internet connections. Caveats: Horne's remarks are prepared text, many recent attacks are only 'thought to have been aided' by AI, and the piece gives no new funding or legislation.