Staff disquiet as Alan Turing Institute faces identity crisis | Hacker News Reader

Staff disquiet as Alan Turing Institute faces identity crisis

theguardian.com

62 pointsglutamate87 comments

Reading between the lines - actually, just reading the lines - it sounds like another organisation that got infested with the kind of people who are apt to ruin organisations, and perhaps an attempt it being made to fumigate it, and they don't like that.

There are plenty of anthropological, sociological, and philosophical arguments that this is inevitable for any organization. Once visionaries in any organization of any size are gone, the priorities of its members are proportionally dominated over time by simple self-preservation.

> Once visionaries in any organization of any size are gone, the priorities of its members are proportionally dominated over time by simple self-preservation.

This. It’s also nigh impossible for new visionaries to succeed in an organization because of that self-preservation of the existing ruling political class. Visionaries show loyalty to the org, not the people, and that makes them a prime target for harassment and cuts as a result.

Smart orgs keep visionaries in charge, but accountable.

And when they can't suppress visionaries within the organization, the ruling class finds other ways to preserve their power. Today the entrenched companies, organizations, and government bureaucracies collaborate to write the laws and regulations and pursue predatory competitive strategies that suppress new competitors and preserve their power.

There are always good reasons for why X law or Y regulation exists or why Z company is given preferences and even subsidies. But the visionaries are undermined all the same.

Without knowing any details, I'm guessing what's happened is the inevitable result that befalls organizations as predicted by Pournelle's Iron Law of Bureaucracy:

Pournelle's Iron Law of Bureaucracy states that in any bureaucratic organization there will be two kinds of people: First, there will be those who are devoted to the goals of the organization. Examples are dedicated classroom teachers in an educational bureaucracy, many of the engineers and launch technicians and scientists at NASA, even some agricultural scientists and advisors in the former Soviet Union collective farming administration. Secondly, there will be those dedicated to the organization itself. Examples are many of the administrators in the education system, many professors of education, many teachers union officials, much of the NASA headquarters staff, etc. The Iron Law states that in every case the second group will gain and keep control of the organization. It will write the rules, and control promotions within the organization.

And as Douglas Adams stated in the problem with ruling axiom, is that under no circumstance should you allow someone who gets themselves in a position to rule, rule.

> under no circumstance should you allow someone who gets themselves in a position to rule, rule.

Unfortunately the opposite is also true, as anyone who's served on a non-profit board with noncommital members knows.

At least some greek states used to assign leadership in a lottery system designed in such a way partially to avoid the issues with this...

https://en.wikipedia.org/wiki/Kleroterion

Also https://en.wikipedia.org/wiki/Sortition

I think a reasonable compromise would be to elect a top 2-3 and then choose randomly between them.

The old election system for one of the Italian city-states was fascinating along these lines - multiple rounds (like, a dozen) alternating between sortition/lot and voting. The benefit, I guess, being that the lottery rounds make strategic voting unworkable, and the voting rounds allow better candidates to be preferred over worse ones.

Lol this is a terrible system because most people do not under any circumstance want to be bothered with wasting their time on this shit.

They get in that position because nobody else wants to do it.

Funny given there was a heated discussion yesterday on the FFMPEG Assembly post[0]. I've been seeing similar heated conversations turn up more regularly. Honestly, I think this is a result of this Iron Law. Arguments between developers devoted to the goals of "the organization" vs those dedicated to "the organization", albeit a bit more abstracted.

Funny, we have similar views about Google search[1], and those days were much better

[0] https://news.ycombinator.com/item?id=44940485

[1] https://www.jerrypournelle.com/archives2/archives2mail/mail4...

This law needs the addition of a third kind of person: one who is neither devoted to the goals of the organization nor the organization itself, but merely wishes to use the organization as a vehicle to push their own social and political beliefs (such as DEI).

I think this might just be a special case of Type II. By latching onto the latest hot issue (like the Linux Foundation getting into Vaccine Passports... or, less subtly, the Firefox organization rebranding as a "global crew of activists"), you get to collect donations from government and public grants related to the issue. And corporate donations, too, because your good "ESG score" transitively applies to your supporters.

They are devoted to themselves.

This is great, I've been looking for a way to express this vague notion I had about organizations. I was thinking along the lines of "nonprofits tend to become generic donation-seeking, rent-seeking entities, keeping their original goals only nominally" but the quote above is a less harsh way to put it.

As a NASA employee, I think this is extremely true. I would bet most of the second group doesn’t or even comprehend that they’re doing it.

I don't think you need to read between the lines. The lede is buried, but explicit:

> This is part of an overhaul dubbed Turing 2.0 under which the institute will focus on three key areas: health, the environment, and defence and security.

They're trying to make the organization into a defense subcontractor (with a few side-projects for image maintenance), and purging anyone who isn't interested in that mission.

I have friends who work there. This is absolutely the answer -- together with the government joint beyond hope that it will be the case that AI solves the UK's economic woes (without ever really understanding what it is)

I’m sure exactly those words were said about Turin

Though its authenticity may be discredited in the modern AI era, many computer scientists still venerate the shroud of Turing.

Yes, we're pretty accustomed to these non-specific accusations of "toxicisity" at this point. It's code for "my pet projects and initiatives have been cut because they produce little of value."

AKA academics.

A lot of academic researchers who couldn’t get jobs elsewhere ended up there.

The org is fundamentally dysfunctional according to an insider I spoke to. They blamed leadership.

They should have isolated it from academia - no hires from universities allowed, only hiring people who had previously worked in industry for the last few years.

You can get a lot of valuable work out of "unmanageable" academics if you figure out how to manage them. That's most of the secret of early Google's success. If we knew anything about the NSA it'd probably be how they work too.

Now this may be difficult in the UK because all jobs pay two quid a year, you have to live in a closet somewhere with a name like Pennyfarthing-upon-Longbottoms, and you can't get air conditioning without permission from the king.

Renaissance Technologies as well. One of the top quant trading firms, with a bit of a mythical aura around them, was specifically seeded with academics, and it also has an academic-style internal environment (vs other quant funds, which also hire academics but leave most of the culture at the door).

Also seeded with NSA talent, with a founder who worked in academia and codebreaking, just to make your comment even more relevant.

The way you "manage" them is you water down their impact on the organization with a bunch of people who aren't academics.

Eh no not really. Google's early success came from PhD dropouts (Larry Page & Sergey Brin), combined with people like Jeff Dean and Sanjay Ghemawat (formerly at DEC), and the lesser known but no less critical Eric Veach (formerly Pixar).

Those are the people behind the core innovations that made Google so powerful: PageRank, MapReduce, BigTable, the indexing pipeline, protobufs/Stubby, GFS, AdWords, AdSense and the use of ML in both (iirc Veach was a key reason the ads system worked so well).

So none of the names I associate with early Google were academics. They all came from industry.

> In March last year more than 180 staff wrote a letter to leadership expressing “serious concerns” about the organisation’s approach to diversity after it appointed four men to senior roles.

If you change the sex and it becomes a sexist statement, it was always a sexist statement.

> “The ATI brand is well recognised internationally,” says Dame Wendy Hall, a professor of computer science at the University of Southampton and the co-chair of a 2017 government AI review. “If it ceases to be the national institute for AI and data science then we are at risk of weakening our international leadership in AI.”

'our [UK?] international leadership in AI' -> citation needed?

DeepMind was founded in London, UK and still headquartered there, and is one of the leaders in the field..

Notable well-known things from DeepMind are AlphaGo (the first time a computer beat a world champion at Go), AlphaFold (resulting in a Nobel prize). Gemini (LLM, a variant of which is used in Google search results) and Gemma (open-weights LLMs).

They were acquired by Google, so you could argue they aren't centred in the UK any more, but I still think they qualify as international leadership in AI coming from the UK.

> They were acquired by Google, so you could argue they aren't centred in the UK any more [..]

They've been acquired. There's nothing left to argue.

I was reading earlier today about how Kuka AG, the German mechanical engineering company, was sold to a Chinese investor in 2016. A fascinating story:

https://en.wikipedia.org/wiki/KUKA

"ATI has recently notified about 50 staff – or approximately 10% of its workforce – that they are at risk of redundancy and is shutting down projects related to online safety, tackling the housing crisis and reducing health inequality."

Why did they have projects in those areas at all?

The UK has a bunch of serious issues.

But actually fixing them would require hard decisions.

So our government wants to look like it is frantically working to achieve these things (fix our housing crisis etc). But absolutely not making the changes needed.

So they announce things like the Turing institute to look busy. Then dump weird requirements on it to look like they’re doing things. Then defund it (because it was never meant to be a real thing, just a donkey to pin press releases on).

See also the “Spaceports” we built, our “massive breakthroughs” in fusion energy and SMRs, Heathrow extension etc…

>In March last year more than 180 staff wrote a letter to leadership expressing “serious concerns” about the organisation’s approach to diversity after it appointed four men to senior roles.

These are not serious people interested in cutting edge AI research.

An earlier discussion on HN about this issue.

What went wrong with the Alan Turing Institute? (April 2024): https://news.ycombinator.com/item?id=43493313

Instead of founding this institute and spending however much they did on it they should have just protected DeepMind and not allowed it to have been sold to Google

Lots of talk about Alan Turing's "legacy" being at stake, "cornerstones", and such - when the story admits that the Institute is only 11 years old. And that the gov't cut & pasted Turing's name onto the front door, 60 years after his death.

And trying to read the article - the narcissistic Dilbert-speak never stops.

Theory: The ATI was founded purely as an exercise in pretentious political posturing. And even on Day 1, it was so badly infested with self-serving third-raters that there never was any chance of it succeeding.

And uhhh... while us computer science types know him for vast contributions to our field... the biggest contribution that the UK Government likely cared about at the time was a massive contribution to (inter)national security and defence!

Right? I had a picture of Turing on my wall long before this institute existed and it will hang there long after no matter what these fools do.

> "In March last year more than 180 staff wrote a letter to leadership expressing “serious concerns” about the organisation’s approach to diversity after it appointed four men to senior roles"

This is part of the "identity crisis"?

Gods forbid that these four men might have been best qualified for the job, or are we reducing suitability to box-ticking again?

When it was founded in 2014 it was criticized as yet another glass building in London (technically a floor in this case) in a very prestigious location. And indeed as you could a lot of the funding went into the building, maintenance, events/catering and you could see random freeloaders loosely associated with the Institute using the space as a free coworking space. I think since the beginning, the PhD funding was great idea as you could do your research towards current issues, somewhat outside of the usual rusty academic echo chambers. But the fact that you were supposed to commute to the central London location, a lot of the grant went on train tickets or accommodation. As an early LLM adopter / practitioner, I went there for some sessions on AI Ethics and such and did not see that it was worth the millions pumped into the institution as we saw that whatever Captain Obvious insights (guardrails, data protection etc) came out of the Institute were completely ignored by the US giants. The current political twist toward practical applications in defence might actually be good for the institute as they will actually be able to practice some applied science but frankly I don't have much hope that my tax payer money is being put to good use here - it's always been a desperate scream for relevance and there's more and more of this action free nonsense coming from the government, like the recent OpenAI memorandum https://www.gov.uk/government/publications/memorandum-of-und...

This is sort of the problem with nonprofits and NGOs generally - they have bad incentives, are easily corrupted, and attract people that don't create any value.

It's the communist form of a company and shares similar failures. IMO we're better off just not having them for the most part.

The purpose of most non-profits would not make any sense as a company, and is not easily amenable to being measured in terms of 'value'. What value is feeding poor people? Technically negative, as that food could have been sold for a profit. But that is of course not a useful metric here.

I agree that it is difficult to align incentives for non-profits, but turning them into companies would simply add a profit motive and an obligation to shareholders on top of those difficult-to-align incentives.

The people that non-profits are accountable to (the poor, minorities, etc) are generally powerless vis-à-vis those non-profits, and there is a perpetual risk of corruption arising from that effective lack of accountability. The paying customers of a business are relatively much more powerful vis-à-vis that business. If Gmail upsets you, you switch to Fastmail; if your soup kitchen upsets you, you... what? Don't eat?

This stuff is very, very hard, and something I'm sceptical will ever be solved, least of all here on HN.

> What value is feeding poor people? Technically negative, as that food could have been sold for a profit. But that is of course not a useful metric here.

In the UK, poor people can afford food that's sold at a profit. They're given free money.

In SF non profits actively make the problems they’re attempting (in theory) to solve worse, while enriching themselves at (often gratuitous) cost to the tax payer.

I understand your argument, but in practice SF would benefit from these shutting down.

At a minimum they shouldn’t be tax payer funded. Even philanthropic non-profits are often funded by children (or spouses) via inherited wealth from the people that actually built things. This then funds actively harmful policy and orgs because the people throwing money around have no idea how to achieve what they want and there is a class of specialized NGO vultures that go after the money while accomplishing nothing.

This is anecdotal, but what I’ve seen of friends that work at these places is wild. People stealing money, extreme entitlement, stupid policy, enormous waste, no accountability. It’s bad.

> In SF non profits actively make the problems they’re attempting (in theory) to solve worse, while enriching themselves at (often gratuitous) cost to the tax payer.

Yes, that's why FAANG keeps losing employees to soup kitchens. The fabulous pay.

> This is anecdotal, but what I’ve seen of friends that work at these places is wild. People stealing money, extreme entitlement, stupid policy, enormous waste, no accountability. It’s bad.

Real issues. I don't think I ever said that non-profits work well; they don't. My problem is with the idea that there's some simple alternative that we could whip up in VS Code, some clever business structure that somehow makes the organisation immune from basic incentives.

Non-profits become bloated and ineffective in exactly the same way as monopolies, and for exactly the same reason - their 'customers' are powerless against them. This is a fundamental issue of power relations, not something someone designed to work this way because they thought starting a soup kitchen is a great get-rich-quick scheme (in the Bay Area, no less. No other way to get rich quick there!).

There's an incredible amount of naïveté in tech criticism of non-profits, and people who end up hurt by that are people in need.

How are non-profits "accountable" to the poor? Are the poor going to threaten to stop giving them donations if they don't shape up? You seem to contradict yourself when you say that indeed if the soup kitchen upsets you then you can't do anything about it. Thus the soup kitchen isn't accountable to the poor who use it.

I think non-profits are accountable to their donors but the problem with charity is that the donors are giving money mostly to be seen to give money. They rarely care much about outcomes. Indeed if the non-profit actually solved the problem they were set up to tackle the donors would have a problem as now they'd need to find a new cause to demonstrate their philanthropic loveliness with.

And yeah one can argue that feeding poor people is of little value; that's the whole idea behind the parable of teaching a man to fish. Translated into modern terms, the right thing to do in terms of value creation is make poor people richer, not give them free food. Then they can feed themselves and much more. It's of course a harder problem but much more valuable to solve.

> How are non-profits "accountable" to the poor?

They're not, in any effective way. Re-read my post. All attempts to make non-profits accountable to someone else - government, donors - are an attempt to work around the fact that the only real stakeholders in a soup kitchen are completely powerless, and wouldn't it be nice if someone powerful could exert accountability? It would! And yet it doesn't seem to work in practice, perhaps because the homeless' interest is in soup, and the donor's interest is in plaques with their names on them, and the two aren't the same thing. It's often donor pressures that reduce non-profit effectiveness.

> And yeah one can argue that feeding poor people is of little value; that's the whole idea behind the parable of teaching a man to fish. Translated into modern terms, the right thing to do in terms of value creation is make poor people richer, not give them free food. Then they can feed themselves and much more. It's of course a harder problem but much more valuable to solve.

It is obvious to the point of complete absurdity that making poor people richer would alleviate poverty. Meanwhile, while you work on the small, pesky matter of solving poverty, people are hungry right now. Are you proposing we shut down soup kitchens until your plucky effort to solve all poverty succeeds? Because if you're not, we're back to where we started.

Posts are like these are precisely why HN isn't solving poverty. "Wouldn't it be nice if the poor weren't poor? Maybe we should teach them how not to be poor? That's surely never been tried before."

I was/am confused by this part then: "The people that non-profits are accountable to (the poor, minorities, etc)". But as you say, they aren't accountable to the poor or minorities really.

In a world with unlimited time and resources, obviously we'd want to do both soup kitchens and other anti-poverty initiatives, but opportunity costs are real. There's limited resources, time and attention to go around. The people spending money on soup kitchens aren't spending money on other things.

In most societies there's a lot of low hanging fruit for reducing (absolute) poverty. Lots of things we know work well to create wealth and reduce poverty get ignored. For example, maybe if taxes were lower fewer soup kitchens would get funded, but fewer people would find themselves needing them to begin with - a win. We know that small state libertarianism creates wealth, so initiatives to address poverty often end up creating the issues they're trying to solve.

Given that, if you have three people and limited time/money, is it better to run a soup kitchen or lobby against poverty-creating government policies? You can't necessarily do both.

Substitute "run education schemes" for lobbying if you prefer. Same tradeoffs apply.

> I was/am confused by this part then: "The people that non-profits are accountable to (the poor, minorities, etc)". But as you say, they aren't accountable to the poor or minorities really.

That's fair. I wrote the post quickly, wanted to avoid words like 'stakeholder' and ended up inadvertently overloading the word 'accountable'. What I meant to say is that a soup kitchen is effectively in a monopoly position vis-à-vis its 'customers', its 'customers' are largely powerless against it, and that creates bad incentives. Trying to engineer clever business structures for the soup kitchen does not remove said incentives.

> (tail)

I am very sympathetic to both libertarianism and soup kitchens, so I'm going to be a somewhat idiosyncratic defender of the latter, but I think the moral hazard argument against charity is a relatively weak one. In reality, people in need of aid divide largely into two groups: (a) people in temporary straits, often through external factors (e.g. fleeing an abusive relationship with nothing but the shirts on their backs), and (b) people with persistent mental health issues. Neither are really groups who would benefit from education on how not to be poor - it would be patronising to the former and wasted on the latter. There is widespread societal consensus that people in situations such as the above probably shouldn't starve to death, which I would posit is a good thing.

What international leadership in AI does the UK have? any models produced?

DeepMind was founded and is still headquartered in London.

"DeepMind Technologies Limited, trading as Google DeepMind or simply DeepMind, is a British–American artificial intelligence research laboratory which serves as a subsidiary of Alphabet Inc. Founded in the UK in 2010, it was acquired by Google in 2014 and merged with Google AI's Google Brain division to become Google DeepMind in April 2023"

Q: Is the HQ nominally being in London at all relevant given it was acquired by Alphabet/Google? I'm sure the accountants have the tax status all sorted by now...

a lot of the researchers are still in london

I understand that iPhones are assembled in China, India and Vietnam. Would those countries issue press releases on "their leadership in iPhones"?

China, India and Vietnam absolutely lead in manufacturing iphones, yes.

in manufacturing <-- did you notice this bit? :)

I was responding to the quote from Dame Wendy Hall claiming that that UK [has] "international leadership in AI"

But in this case Deepmind's assembly (plugging in datacenter cables, etc.) is largely done in the US.

I'm sure representatives of those countries love to say so, especially when talking to third parties about their expertise in manufacturing: "Yes, here in Zhengzhou, CN we're leaders in electronics manufacturing - the iPhone is assembled here at Foxconn!"

However, Apple (headquartered in the US) loves to issue press releases describing how their products are "Designed by Apple in California[, USA]" even though a lot of work in the manufacturing, the software, and the design of subcomponents (or major components, I don't know how Apple is organized internally) are done in China, India and Vietnam as you listed.

I'd argue that in the same way that Shenzen and Zhengzhou are leaders in electronics assembly because the bulk of the iPhone and other products are built there, regardless of the location of the headquarters of Apple, so to can London claim to be a leader in AI because the researchers for DeepMind are located in London, regardless of who owns the DeepMind brand.

Buying a thing from another country doesn't make your location a leader in that thing.

> "we're leaders in electronics manufacturing"

The UK wasn't claiming to be "leaders _in manufacturing_", they were claiming "international leadership in AI".

As I said elsewhere in the thread, citation needed...

What you're replying to is an analogy, that you seem to completely fail to understand.

Apple's manufacturers don't do any of the work on the software or design. They don't even manufacture the highest value-add components; those are mostly done in Taiwan and Korea.

Also in general Google satellite offices often house the engineers of acquired startups who don't want to move to the mothership. It's not their primary purpose but it's one of the things they use them for.

> in general Google satellite offices often house the engineers of acquired startups who don't want to move to the mothership

Would it be unfair to ask if (in this instance the UK's) satellite country taxpayers are subsidising corporate offices when the overall structures are arranged such that any overall corporation tax payable will be paid in the lowest-possible jurisdiction?

See - for instance - https://en.wikipedia.org/wiki/Corporation_tax_in_the_Republi...

It's not just the HQ, the only AI researcher I know personally is an American who moved to London to work on AI with DeepMind well after the acquisition.

The registered HQ and a large research center are in London, but ownership, executive control, substantial staffing, a big fraction of the training/serving compute, and the commercialization pathway run through Alphabet's U.S. operations, so the work is, in practical and legal senses, U.S.-based...

"As part of a wider group reorganisation, the Company distributed intellectual property assets which had a nil book value to another group undertaking on 31 October 2019."

Honestly, claiming DeepMind is still some scrappy London-based startup is quite unfortunate :/

+1 Europe (especially) and the UK largely don't matter - it's a battle between the US and China, the gap will only grow wider and faster than it already has (and it's already getting really noticeable).

I attribute it mostly to a cultural problem and I don't think they can fix their politics from the downward spiral they're on. It's why they have a number that rounds to zero of billion dollar software companies and why all their ambitious people do their best to get to the US.

> The registered HQ and a large research center are in London ...

> so the work is, in practical and legal senses, U.S.-based...

These two statements literally contradict each other in both cases.

> These two statements literally contradict each other in both cases

Welcome to how multinational corporate entities structure their tax affairs!