GCHQ has issued an urgent warning: the use of artificial intelligence to simulate deceased Ukrainian soldiers for propaganda and psychological operations is an 'unprecedented ethical breach' that could destabilise the already fragile moral framework of modern warfare. According to sources within the British intelligence community, Russian-aligned groups have been employing generative AI models to create hyper-realistic video and audio messages from fallen Ukrainian troops, designed to demoralise their families and former comrades.
'This is not just disinformation. This is digital necromancy,' said a senior GCHQ analyst who spoke on condition of anonymity. 'We are seeing deepfakes that capture not only the appearance and voice of deceased soldiers but also their mannerisms and speech patterns. The psychological impact on those who receive these messages is devastating. It weaponises grief.'
The technique, referred to internally as 'digital resurrection', leverages generative adversarial networks trained on vast datasets of personal media, including social media posts, video calls, and military correspondence. By synthesising this data, the AI can generate plausible interactions that the deceased might have had, making it nearly impossible for laypeople to distinguish real from fabricated.
Ukraine’s Cyber Defence Command has already documented over 200 cases of such simulacra being distributed via encrypted messaging apps targeting military families in the Kharkiv and Donetsk regions. In one instance, a mother received a video call from a digital twin of her son, killed in action three weeks prior, pleading for her to 'bring him home'. The distress caused was described as 'severe' by local psychologists.
This development raises profound questions about digital sovereignty and the rights of the dead. Who owns a person’s digital likeness after they die? Should there be legal protections against such misuse? In the UK, the Law Commission is reviewing potential legislative responses, but the pace of technological change threatens to outstrip the rule of law.
'We are entering a realm where the dead no longer have a voice, but their digital ghosts can be made to say anything,' warned Dr. Eleanor Hart, a leading AI ethicist at Oxford. 'This is a dangerous precedent. If we do not establish international norms now, we will see this tactics spread to other conflicts, even to crimes and domestic abuse.'
British intelligence is urging tech companies to implement stronger detection algorithms and to close loopholes that allow these simulations to be created from scraped data. Meanwhile, NATO allies are being briefed on countermeasures, including the use of digital watermarking and behaviour-based anomaly detection in communication networks.
The ethical order of war, already strained by autonomous drones and cyber attacks, now faces a new human-machine boundary. As one GCHQ veteran put it, 'The sanctity of the dead is not a technical problem. It is a test of our humanity.'
