“President Macron’s deepfake might seem like harmless fun to promote the AI Summit in Paris, but it is not in general a good thing,” says Paul McKay, VP and principal analyst at technology consultancy Forrester. “Normalising deepfakes in this way should not be encouraged as it continues the difficulty with telling what is real and what isn’t, and is ultimately helping to establish what is fact from fiction.” Dr Richard Little from Salford Business School also warns of the risk of “normalising” deepfakes videos, “both on social media and by scammers.” He adds: “It is great to bring attention to this threat, but doing it in a way that shows how easy it is to create deepfakes risks their wider adoption.” Prof Philip Howard, president of the International Panel on the Information Environment, says AI is increasingly being used “in innovative, sometimes playful ways, as demonstrated by President Macron’s recent video.” But he adds: “These kinds of videos are often released when the guidelines on public communication are not clear.” There has been some debate in the French media over whether Macron should be trivialising deepfake videos when they can be used for harm. In October, Macron told Variety deepfakes “can disinform, which can upset our democracies.” He said they should be regulated “by imposing responsibility on the people who disseminate this content to moderate it.” The EU’s newly implemented AI Act, which regulates the use of artificial intelligence, has faced criticism at the Summit for stifling innovation. The bloc also unveiled plans for a Europe-wide, open source AI model with a budget of €37.4m (£31.1m). A global declaration of shared AI goals and ethical responsibilities is due to be revealed at the end of the Summit tomorrow. So far the US and the UK have both declined to say whether they will sign it.