( UK ) Prime Minister sets out blueprint to turbocharge AI
57 comments
·January 13, 2025janice1999
voidr
> one that won’t sit back and let opportunities slip through its fingers
Maybe they should fix the economy to attract more AI companies and researchers instead, while they are at it they can fix the police force as well to give confidence to said companies and researchers.
The lack of data centers is not the reason I'm not researching AI at the moment, I can buy the hardware or rent a US data center, the reason is that the UK economy is dead, borrowing is insanely expensive and taxes are so high that most of the profits would go to the government, which would be fine if it would spend it on things that matter, but it's not doing that, it's just buying votes.
gazchop
Worse. Much worse. Healthcare and social care will be involved almost immediately and forced to suck up all the bad ideas and inappropriate uses to keep interest up. People who raise rational objections will be silenced.
graypegg
> Sovereign AI compute, owned and/or allocated by the public sector, will enable the UK to quickly and independently allocate compute to national priorities. [...] Sovereign AI compute will almost certainly be the smallest component of the UK’s overall compute portfolio.
> NB: this review has not considered the requirements of non-AI high-performance computing, for which there is already a well-established case, including the need to deliver an exascale capability. Government should seek to resolve this as soon as possible, noting that these systems will play a crucial role in supporting AI science and research.
What an interesting addendum to add to that. The government, or at least who ever is writing this report, seems to imagine "AI" as a uniform and distinct process, where you turn on the AI machine, electricity goes in and AI comes out. "Can't use a non-AI machine. It doesn't make AI." The government should already be investing in computational resources for a huge number of government projects and services. The N.B. makes a good point.
The fact someone needed to add this note, and it made it into the final public document, written in that tone, seems like this isn't really a plan, more so just a vibes-based promise for innovation from 1 team in the government.
dkdbejwi383
> The government, or at least who ever is writing this report, seems to imagine "AI" as a uniform and distinct process, where you turn on the AI machine, electricity goes in and AI comes out
This is how a large chunk of the general public think of "AI" as well. They're just waiting for someone to flip the AI switch or add a bunch of `from ai import magically_do_everything` to products and services they use.
fredley
Don't blame them, this is how products are marketed and sold now, e.g. AI washing machines, AI microwaves, AI features in cars, AI home lighting. Of course, most of these things don't have anything a software engineer or computer scientist might reasonably call AI in - they're just using sensor data and (conventional) algorithms, the same way they have for decades, but AI is the buzzword now.
segasaturn
> seems like this isn't really a plan, more so just a vibes-based promise for innovation
To be fair, that description lines up with a lot of what I hear from other AI startups.
elric
> The government should already be investing in computational resources for a huge number of government projects and services
From what I see, governments are massively shifting to Azure and AWS (moreso the former).
DrScientist
As long as you have decent utilisation then buying and running your own hardware is typically a lot cheaper than using AWS etc al.
Cloud shines for dynamic loads, global resilience etc - but if you just want to run a lot of computational jobs for a long time, then running your own kit is likely to be much much cheaper.
The other advantage of your own kit is you can guarantee it will be available when you want it, and you can optimise the stack for your own applications.
There is a reason something like 3/4 of Amazons profits come from AWS and not selling books et al.
elric
> As long as you have decent utilisation then buying and running your own hardware is typically a lot cheaper than using AWS etc al.
The cynic in me suspects that some government consultant (I used to be one) is making a lot of money from peddling AWS/Azure.
graypegg
It does make me wonder if some nationalized utility like "British Compute" or something would be a good idea. They could over-allocate and sell the unused resources to the private sector, AWS style. Could bring long term costs down even further. (With bonus political levers to pull, like giving British owned companies special rates)
milliams
They're already investing in non-cloud GPU-heavy supercomputers designed for AI research (i.e. into- and using-AI). For example the Isambard-AI supercomputer at Bristol University (https://www.bristol.ac.uk/news/2023/november/supercomputer-a...) and this report seems to be talking about doing a lot more of that - in addition to non-"AI".
Disclaimer - I work at Bristol on Isambard-AI
dkdbejwi383
> feed AI through cameras to spot potholes and help improve roads.
This example only makes sense if we're in a position where all the potholes we know about are fixed, and all of the pothole fixing masterworkers are sitting idle, waiting to leap into action at the next report.
But the reality is there are many, many potholes that are known about, but we can't/won't fix them due to things like budgets and staffing constraints.
fredley
This is absolutely insane. If they want to know where potholes are, throw up a basic site with a map for reporting, people are very motivated by issues on the roads. Or just ask Google to share Waze data - many potholes on major roads are tagged there. "Feeding AI through cameras" makes so little sense it's laughable, in the most depressing way.
As you've pointed out, I might let it slide if issues were actually being addressed, but they aren't. There are potholes that have been sitting in major city roads where I live for years, and nearly all the street markings around the whole city have completely eroded - junctions are becoming scarier as nobody knows what lane to be in. But I'm sure throwing money at AI can solve these problems.
cameronh90
Such a site already exists: fixmystreet.com. Most councils already use it (or an instanced version of it), and every time I've found an issue in public, it has already been reported on there. Often multiple times.
The issue is, as you say, the council doesn't have the budget to actually fix the reported problems.
gpderetta
The catch is that if your car is damaged by a pothole you have a chance to claim damages from the council only if you can prove that the council knew about the pothole. So even if most reports do not result in an action, they still have some values and are an incentive for the council to fix the road.
salmonlogs
I've submitted many reports through similar apps to my local council and find 25% of them are even acknowledged, 10% at most are "fixed" and the fixes are worthless.
A pothole I reported last October was "fixed" in December, and has already opened up again after the ineffective repair failed to take.
DrScientist
Indeed.
The problem doesn't lie in detection - the problem lies in poor road maintenance - which is presumably driven by various effects of cost cutting - including outsourcing to the lowest cost providers who have an interest in repeat business rather than a job well done.
andtheboat
Have you tried killing all the poor?
autonomousErwin
What about changing the hyper-parameters and killing all the poor?
scrlk
> We can learn from the US’s and EU’s approach - delivering the dynamism, flexibility and long-term stability that we know businesses want.
At least for AI regulation, the EU is the last place I'd want the UK to learn from.
ljm
As opposed to the US, which is famous for its hands-on, consumer-friendly attitude to business?
The challenge for the UK is actually building this stuff to a non-zero level of competence. The usual suspects in the IT consulting industry must be collectively having seizures over the prospect of being paid another 7 figure sum to deliver absolute slop several years behind schedule. Sunak beating himself up over the missed opportunity to inject another billion in cash to InfoSys.
scrlk
I'm not saying all EU regulations are terrible. The impact of their AI regulations have resulted in Europeans either receiving new models or features later than other regions, or not receiving them at all.
There has to be a balance between a complete free for all and potentially snuffing out Europe's ability to compete.
will_critchlow
I hoped as I read that sentence that "learning" could involve learning what not to do as well as learning what to do.
scrlk
Good point.
I note that they've announced "work starting on a brand new supercomputer", so this is potentially backtracking on a very poor decision made last August to cancel a new exascale supercomputer at the University of Edinburgh. (https://www.bbc.co.uk/news/articles/cyx5x44vnyeo)
DrScientist
In terms of streamlining paperwork with AI - I'd first look at whether the paper work is useful in the first place, before spending lots of money trying to automate it.
For example in the NHS, the 'market' reforms of successive governments have led to the ridiculous scenarios where hospital doctors can't simple book a follow-up appointment to check on a patient in a few weeks time - they have to do all the discharge paperwork and ask the patient to get a referral from their GP back to the service - as that's how various bit's of the org 'get paid'.
graypegg
To some extent, there is actually an incentive to have low-value paperwork in a health system. There's only so many resources, and adding friction to the process is a form of "how serious is your issue, really?" triage. It obviously falls apart where the doctor is doing the same thing they normally would, just now with more paperwork, but it may make them think twice about doing two blood tests when they have a hunch the first one is the most relevant.
Of course, that sucks for every metric of a health system that actually matters (outcomes), but this report could reflect one hand of the government not talking to the other.
DrScientist
You seem to be under the assumption that Doctors aren't interested in triaging, and the optimal allocation of resources. Of course they are - they want to be spending their time the most effectively in helping patients.
They don't need a paperwork hurdle to decide whether a blood test is needed.
Another reform that causes endless problems is the idea of patient choice/patient voice - which means vocal patients who are good at playing the system, and who are prepared to complain, suck up a disproportionate amounts of the budget - as it's easier to spend unnecessary time than it is dealing with the complaint paperwork.
Note the common theme of the above two pathologies - a systemic lack of trust in professionals ( in this case doctors ) in being able to do a good job - by people who aren't said professionals....
graypegg
No I do agree with that! I assume doctors are working in the best interest of their patients. That includes not running arbitrary tests. But I can also imagine a bureaucrat thinking an explanation for each test to be promoting "transparency". (Read: discourage spending as transparency in this case means you're being audited, reflecting that distrust of professionals from non-professionals.)
Not saying it's a good thing! Just that one team has been told the forms are important by another team, so they pitch automating the filling of the form, rather than the obvious solution of: no form.
The patient choice aspect is interesting. These processes becoming a game played by people wanting to take advantage of the system is another aspect where this form-heavy process fails, but a bureaucrat could argue that it still (however badly) discourages overconsumption of the system from patients.
Edit, I should mention: I am Canadian, though we have similar issues with our provincial health systems. If there is a specific British/NHS aspect to this I am missing, that will be why.
uncertainrhymes
The tone of this sounds like it was written by non-bureaucrats trying to score political points. I get that it's a press release, but it sounds like a stump speech.
"Today’s plan mainlines AI into the veins of this enterprising nation – revolutionising our public services and putting more money in people’s back pockets."
Is a heroin analogy really what they are going for, and then straight into some imaginary cost savings?
I know it is fashionable to wonder if something was written by AI. I don't think this is, but some human went to great lengths to sound borderline unhinged.
graypegg
I'm starting to notice that the main way for me to pick up on AI written content is the tone! No LLM will give you a heroin analogy unprompted. The AI written version of this press release is a dry desert of weak adjectives, but a bureaucrat that's been asked to "add a little more... you know, zhuzh it up" will absolutely transport the reader into a lush Afghani poppyfield.
UK-AL
I like AI for specific tasks its suited to and the tech involved. But i'm skeptical it can create widespread massive productivity improvements. I think its bit oversold on that.
We seem to lack people to fix pot holes, not report them. We need carers to support old people. AI seems to be mostly good at writing meeting minutes, reports and generating slightly incorrect content/code.
Happy to be proven wrong though.
nsteel
> AI is already being used across the UK. It is being used in hospitals up and down the country to deliver better, faster, and smarter care: spotting pain levels for people who can’t speak, diagnosing breast cancer quicker, and getting people discharged quicker. This is already helping deliver the government’s mission to build an NHS fit for the future.
I find this really hard to believe. My brother, who is a practising ward doctor by trade with almost zero software experience, recently did a sabbatical related to cleaning patient notes data for use in training. He said it was a hopeless mess and they had absolutely nothing. The work he did went nowhere. I appreciate the hospital trusts are different and isolated in some respects, but the idea they're doing anything with AI is a joke when they can't even do the basics (as anyone who's used NHS digital services will testify to).
Does anyone have any experience that actually agrees with this press release?
teamonkey
The press conference earlier today implied the government was spending on AI, but this suggests the opposite: they've secured private investment to the tune of £14bn and thousands of UK jobs.
I can't see any mention of direct costs to the UK taxpayer, except for needing to include AI in the NHS and civil services in various vague ways.
There are no details on what laws or regulations they plan to change in exchange for that investment, though guessing from the public questionnaire a few weeks ago, they will likely relax copyright on private artworks so it can be used with AI generation by default. Tax breaks are a given but it’s not clear how much or if it applies to anyone working in the sector.
This all means nothing if the AI bubble pops, forcing those companies to collapse, laying off thousands and wasting everyone's time (and rights).
ChrisArchitect
Some more earlier:
'Mainlined into UK's veins': Labour announces public rollout of AI
ryanmetz
They’re gonna create a competitor to OpenAI using the British Civil Service and government spending. Yeah, ok. Best of luck to them.
fmajid
Using half of what Meta, the smallest of the FAANG, spends in a year alone on AI datacenters.
indy
"Sure OpenAI is ok, but what if we created a competitor with the imagination, innovation and work ethic of the British Civil Service, all on a fraction of the budget"
>the AI industry needs a government that is on their side, one that won’t sit back and let opportunities slip through its fingers. And in a world of fierce competition, we cannot stand by. We must move fast and take action to win the global race.
This sounds like desperation more than anything else. I think a rush to use AI in public services without thorough planning and a critical mindset will inevitability lead to another Post Office or Robodebt scandal [2].
[1] https://www.computerweekly.com/feature/Post-Office-Horizon-s... [2] https://www.bbc.com/news/world-australia-66130105