Health secretary Yvette Cooper has accepted almost £500,000 from a businessman with commercial ties to the NHS, an analysis by the BMJ has found.
Cooper, who was appointed to the new role by prime minister Andy Burnham on 20 July, received the donations from medical recruitment entrepreneur Peter Hearn and two of his firms.
According to her parliamentary financial disclosures, she received £248,619 directly from Hearn, as well as a further £220,623 over the past ten years from two companies: MPM Connect and OPD Group.
Hearn, who died recently, held a significant stake in Odgers Berndtson, a recruitment firm that worked extensively with the NHS.
Odgers has also acted as a recruitment partner for a number of senior NHS appointments. It headhunted former TalkTalk boss Dido Harding to oversee the controversial Test and Trace programme during the Covid-19 pandemic.
Cooper's donations were declared and there is no suggestion that she breached parliamentary rules. But doctors and campaigners have raised concerns about the perception of a conflict of interest.
“NHS doctors are prohibited from accepting small gifts from pharmaceutical companies, such as a pen or a notepad, because of the risk of the perception of influence.” Steve Taylor of the Doctors’ Association UK told the BMJ. “Ministers responsible for the NHS should be held to at least the same standard.”
“Political donations from individuals or companies with commercial ties to the NHS... inevitably raise questions about whose interests are being served,” said GP Emily Ball, director of the anti-privatisation group EveryDoctor.
“Mr Hearn kindly donated money to fund Yvette’s office staff, which has all been fully declared and is compliant with all the rules,” a spokesperson for Cooper told the BMJ.
Hearn was a major Labour party donor, with former health secretary Wes Streeting also receiving more than £285,000 from the mogul and his companies between February 2019 and February 2026.
Novara Media approached Odgers Berndtson for comment.
Just saw the flag fly over Varanasi as India celebrates its historic win in the G20 summit, pushing climate action and tech diplomacy forward. 🎉🌏 #IndiaRising#TechDiplomacy#ClimateAction
I use other people's open source without a second of hesitation, and opening my own took me a long road. Meanwhile my customers need open components to close their own doors. The two things turn out to be the same problem. https://t.co/4VNuvA57pP
Some random take at the end of the day...
The restructuring changed my job more than I expected.
It has been a little over a week since we restructured the team, and I am still noticing how much it has shifted not just how we work, but how I personally spend my time.
When I first shared the decision, I think most people naturally assumed the biggest change would be organizational, mainly the size of the team or how responsibilities were redistributed. I thought the same at a high level. In reality, that was not where the real change showed up.
The most significant shift was in my own role.
Going into this, I expected to spend more time building. The assumption was simple: if AI makes engineering more efficient, then fewer people should mean more time writing software and moving faster on implementation. That was the mental model I had.
But that is not what actually happened.
Instead, I slowly found myself stepping away from hands-on implementation and moving toward work that is much harder to define. My day is no longer centered around writing code. In fact, I now write significantly less of it than I did just a few months ago.
Most of my time is now spent reading through specifications, reviewing AI-generated outputs, checking assumptions that were made upstream, and making judgment calls on whether something is actually ready to move forward.
At first, I assumed this was just a temporary phase. I thought that once things stabilized after the restructuring, I would naturally drift back into my usual engineering rhythm.
But that never really happened.
As AI became capable of producing more complete pieces of work, my role gradually shifted without any explicit decision. I stopped being primarily a builder and started becoming more of an evaluator, someone who is responsible for deciding what is correct, what is acceptable, and what should be improved before it reaches production.
On paper, that might sound like a small or even subtle change. In practice, it is not subtle at all. It changes the nature of the work completely.
When you are writing code, feedback is immediate. You know within minutes whether something works or not. The loop is tight, and the signal is clear.
Judgment work is very different. The feedback loop is delayed and often indirect. You only really understand the quality of a decision days or even weeks later, when real users interact with the system or when edge cases start to surface in production.
What surprised me most is that I did not expect reviewing to become such a central part of my job. Yet over the past week, that has been the most consistent pattern I have observed in how I actually spend my time.
And that realization naturally led me to something else I have been thinking about.
The bottleneck in our engineering process is no longer how fast we can write software.
It is something else entirely.
I will write about that tomorrow.
PrivOS lets teams tag AI agents directly in chat to execute tasks using files already in the same room, so context never has to be rebuilt in a separate tool.
How much overhead does your team carry moving between different systems today? 💬
#EnterpriseAI#UnifiedWorkspace
Heatwave gripping India's northern states 🥵. Water scarcity & health concerns are rising. Really hoping for relief soon & for communities to get the support they need. #IndiaHeatwave#ClimateChange