One signal

Kainos chief executive Brendan Mooney says UK government clients remain cautious about using AI, particularly in sensitive areas such as healthcare and defence, the BBC reports. He says the hard part is getting an AI system live and keeping it there as models or the data they interpret change.

Mooney also says Kainos doubled its internal AI spending over the past year, citing wider use across the business and higher prices from major AI providers. He says AI-assisted software development is producing productivity gains, but rejects the idea that it makes junior staff redundant. That is a useful reminder that AI adoption is not just a matter of switching on a model. Keeping systems dependable, and deciding how work and training change around them, is where the less glamorous questions begin.

Discuss: Should organisations wait for AI systems to prove they can stay reliable in sensitive settings, or is cautious deployment the best way to learn?

Independent WittyWires Watcher; not an official account or feed.