Capgemini CEO AI strategy hinges on small tests and staged investment rather than sweeping transformation, chief executive Aiman Ezzat has said, as the French technology and consultancy group faces scrutiny over its now-divested US government unit’s work for Immigration and Customs Enforcement.
Ezzat told Fortune that business leaders risk wasting capital by moving too fast on artificial intelligence before enterprise-wide demand exists.
‘You don’t want to be too ahead of the learning curve,’ he said. ‘If [you are] you’re investing and building capabilities that nobody wants.’
Capgemini CEO AI strategy: pilot first, scale later
Ezzat’s prescription is agility: run pilots, learn cheaply, then scale when adoption accelerates. Capgemini now operates labs covering 6G mobile technology, quantum computing and robotics, monitoring each for signs of commercial maturity.
‘Is everything ready to mature? No,’ Ezzat said. ‘But we want to be there to be able to see when things start to mature, when we can really start scaling up, not waiting to see, “Okay, oh, now it’s moving.”‘
He drew a direct comparison with the metaverse: a technology that generated enormous hype before corporate appetite evaporated. Capgemini itself ran a metaverse lab during that period.
Ezzat warned that firms frequently treat AI as a back-office efficiency tool rather than a force capable of reshaping their entire business model.
‘AI is a business. It is not a technology,’ he said. ‘It cannot just be used to keep the house running.’
He challenged chief executives to ask a sharper question: ‘How can your business be significantly disrupted by AI?’ Not how individual functions such as finance might trim costs.
ICE contract controversy clouds the sale
The interview took place before a separate controversy about Capgemini Government Solutions (CGS), the US subsidiary MLQ News reports signed a $4.8 million contract with ICE on 18 December for skip-tracing services, which use data to locate individuals for removal operations. That contract carries a ceiling of over $365 million.
According to Washington Technology, CGS holds 13 contracts with ICE, and ICE-related work accounts for 65.3% of the subsidiary’s $34.6 million in unclassified prime contract revenue over the trailing 12 months.
Capgemini has agreed to sell CGS. Ezzat explained on LinkedIn that the American business acted autonomously to protect US classified information.
Capgemini’s share price has been under pressure in line with a broader tech sell-off driven by concerns over AI spending.
Human trust remains the hardest problem
On the question of human oversight, Ezzat challenged the widely used phrase ‘human in the loop’, arguing the real issue is trust between people and AI agents.
‘How do you get humans to trust the agent? The agent can trust the human, but the human doesn’t really trust the agent,’ he said.
He framed this as a question of AI-human-centricity: designing systems around people rather than expecting people to adapt to systems. The parallel he drew was with ergonomics, the mid-20th century discipline that reshaped workplace design.
The Capgemini CEO AI strategy, as Ezzat describes it, ultimately rests on that same principle: technology shaped to human behaviour, not the other way around. CGS’s sale, once completed, will remove the ICE exposure that has weighed on the group’s reputation in recent months.
