Snippets 11: US deep state, how Apple is managed, laughable HS2 'evidence' to No10 in 2020, why do OneWeb

... and why Silicon Valley benefits from being so far from Washington

[UPDATED AT BOTTOM, 18/10: fake news & war]

Sorry for delays.

I’ve been stuck going through my statement for the official covid Inquiry with the Inquiry lawyers. It has been painful.

I finally sent it in on Thursday. I give evidence on 31/10.

I will put the statement on my blog, but have been asked not to for the moment (and some is being redacted for ‘national security’ reasons). I’ll comment on some of the statements of others when I’ve read them, and do a post-evidence AMA on it all and record some of the most important things that emerge.

Singapore has already done its inquiry, figured out how to improve, and implemented the lessons. This is itself a pointer to how dire Westminster is. In one short news report there are more sensible sentences from Singapore than you find in entire debates in Parliament. Also interesting that Singapore, like the Vote Leave No10, did better than the EU on procuring vaccines — the German government described the EU approach as a ‘shitshow’. Westminster now has little interest in it all other than the ‘which famous people get blamed for what’ stories. Little will change except, probably, Whitehall will create hundreds of new ‘processes’, more non-jobs, and friction will be even higher when the next crisis hits. Such is life in a country that increasingly resembles the trajectory of the Austro-Hungarian Empire — ever more tragicomic baroque bureaucracy plus ‘the system is running the people’.

I will finish going through TSP comments and reply ASAP with thoughts.

I’ve written a few things over the past 8 weeks or so.

I’ll post some below for subscribers in coming days/weeks…

Below a few things including a graph I found on HS2 while going through documents for the covid inquiry, which sums up a lot…

• A new book on the US deep state

• How Apple is managed — in contrast to Whitehall’s systematic war on expertise and long-term building

• What I predicted on HS2 in Jan 2020 & why — and one of the absurd Jan 2020 ‘evidence for demand’ graphs given to No10 purporting to show that *the entire country* would be either on HS2 or building HS3, HS4 etc

• Bill Gurley’s excellent talk on regulatory capture: Silicon Valley has been so successful ‘because it’s so fucking far away from Washington’ (particularly relevant to those working on the UK AI Taskforce)

• OneWeb and LEO satellites, Starlink, war — why did I push through OneWeb in 2020?

And two twitter lists, one for SW1-pundits and one for builders/creators so you can watch them side by side and judge for yourself whether you should continue watching political news.

More coming soon…

And I’m proofreading the Bismarck chronology which I must post before 25/11…

---

Underground Empire: How America Weaponized the World Economy

I’m reading this, just out.

It tells the story of how America, like Rome, built critical global networks:

• communication including the internet and satellites,

• computation (chips, servers, cloud etc),

• logistics,

• banking and payments (e.g SWIFT (1973), the dollar clearing system, and eurodollar market taking over from the miles of pneumatic tubes built in the City in 19th Century),

• underground bunkers of many kinds,

• supply chain infrastructure,

• transport.

In the 1960s Citibank hired teams of locals in Brazil to keep dialling the phone to try to grab open international call connections. In 1988 the first transatlantic fiber-optic cable was laid, as part of the R&D program run by the legendary Bell Labs (9 Nobels, 5 Turing Medals).

American forces preserve security for these networks. The Navy patrols fiber optic cables. Agencies such as the NSA and NRO patrol the internet and spy from space. Critical hubs such as Ashburn, where data centres cluster, are protected by the FBI.

These world-spanning systems were largely built by private companies pursuing profits just as former empires grew with soldiers following merchants.

America’s control of these networks wasn’t scrutinised much after the collapse of the Soviet Empire in 1991. Countries joined global systems for the internet, finance, and supply chains. Prominent journalists like Thomas Friedman hit the lecture circuit and studios declaring the world was ‘flat’ and globalisation unstoppable and wonderful. Bill Clinton told China trying to control information was like nailing Jell-O to the wall.

After 9/11 America started using these networks to fight terrorism. They recorded vast numbers of phone calls. They demanded access to SWIFT messages and got it — with SWIFT executives keeping Brussels in the dark and Brussels officials happy to be kept in the dark (much less concerned about ‘international law’ than our naive government lawyers tell our naive media). Control of SWIFT and dollar clearing were used to coerce other powers. Secondary sanctions forced foreign companies into line. Foreign banks were forced by threats to comply with US demands even if they technically weren’t required to do so — the threat of being blacklisted from the US system was too disastrous to resist. In the old days the US did not try to use the dollar-clearing system to coerce countries, fearing it would encourage a shift away from using the dollar. But they withdrew exceptions to sanctions regimes and stopped foreign banks using the dollar-clearing system for purchases of Iranian oil. The US Treasury became a player in security and intelligence affairs having resisted this for decades. Their officials flew around the world skipping other officials and talking direct to banks: if you deal with Iran, we’ll kneecap you. SWIFT also excluded Iran, an unprecedented step.

The Snowden leaks provoked global discussion of America’s power over critical networks. China started building its own rival networks with companies like Huawei. America took its eye off the ball and quickly the global market for some technologies had shrunk to Ericsson, Nokia, and Huawei — with the first two financially shaky. America decided to kneecap Huawei and imposed sanctions that made it impossible for allies to stick to plans and contracts. In 2020 the US used “foreign-produced direct product” (FPDP) rules on IP to stop foreign companies providing Huawei and others with US IP (including on semiconductors). And it squeezed TMSC to stop supplying Huawei. And it expanded its campaign to trying to kneecap China generally on advanced semiconductors.

America also started imposing sanctions against Chinese government figures, e.g Hong Kong’s then chief executive Carrie Lam, who then experienced the domino effects of US financial dominance — Chinese banks were reluctant to help because they didn’t want to become targets themselves.

(I was involved in discussions over Huawei, espionage etc in 2020. I agree that the West/Five Eyes had taken its eyes off the ball and Huawei should never have been allowed to spread as it did. UK policy was kneecapped by Cameron-Osborne insisting that potential investment from China trumped security issues — classic short-term Tories. We began work in 2019 on the first serious overhaul of our laws to constrain China, Russia and others. Dangerous foreign activity is far greater than MPs realise. On kneecapping Huawei, note that they just released details of their new chips. It seems that US sanctions have backfired, as some predicted, e.g Dan Wang, Steve Hsu.)

Europe has been highly ambivalent. Macron said that once cooperation became dependence ‘you have become somebody's vassal and you disappear’. But Europe has been so hopeless on technology and defence, and so reluctant to face its failures, that it’s both extremely dependent and extremely resentful. On one hand it claims the moral high ground over ‘respect for privacy’, on the other it quietly acquiesces in institutions handing data to US intelligence that can be analysed and given to EU governments while they maintain their public pretence. On one hand they loudly pledge they will build capabilities to reduce dependence, on the other they can’t build and become more dependent (e.g look at their pathetic AI attempts, where the EU’s relevance now is almost entirely limited to what they can undermine/vandalise via regulation — the EU has ZERO relevant AI companies at the leading edge).

Covid and Ukraine have emphasised the fragility of these global networks and heightened global resistance to American control.

China is, of course, adapting. They realise they have to build alternative systems — financial, technological, logistical — and use diplomacy to attract others.

These dynamics will increasingly demonstrate a fundamental problem for the EU and UK. Because we haven’t and don’t take technology as seriously as America and China, we will increasingly be faced with the choice of letting either China or America control critical things, with no option for national independence.

I’ll write notes as I go and post them later…

---

How Apple is managed much more seriously than Whitehall

A central story SW1 tells to itself about itself is that the permanent civil service is justified by ‘the need for a state to operate long term’. Businesses and others need to operate within consistent systems and the state is supposed to manage these long term systems while other players act more short term.

How Whitehall actually works is incredibly short term, generally far more short term than many bad businesses, never mind great businesses, which politicians and officials think of as ‘short term’ relative to them.

This lies behind the dysfunction we see everywhere from procurement to R&D to the NHS to capital budgets to the multi-decade energy policy/infrastructure fiasco to the HS2 fiasco.**

If interested in how the world’s most valuable company works read this.

The level of seriousness is far beyond anything Westminster can muster — seriousness about people, seriousness about expertise, and seriousness about long-term capabilities. Also — seriousness about holding people responsible if they fail, largely unknown in SW1 which protects senior people from responsibility and at most in extremis blames some junior people.

When Jobs arrived back at Apple, it had a conventional structure for a company of its size and scope. It was divided into business units, each with its own P&L responsibilities. General managers ran the Macintosh products group, the information appliances division, and the server products division, among others. As is often the case with decentralized business units, managers were inclined to fight with one another, over transfer prices in particular. Believing that conventional management had stifled innovation, Jobs, in his first year returning as CEO, laid off the general managers of all the business units (in a single day), put the entire company under one P&L, and combined the disparate functional departments of the business units into one functional organization…

Senior vice presidents are in charge of functions, not products. As was the case with Jobs before him, CEO Tim Cook occupies the only position on the organizational chart where the design, engineering, operations, marketing, and retail of any of Apple’s main products meet. In effect, besides the CEO, the company operates with no conventional general managers: people who control an entire process from product development through sales and are judged according to a P&L statement…

Business history and organizational theory make the case that as entrepreneurial firms grow large and complex, they must shift from a functional to a multidivisional structure to align accountability and control and prevent the congestion that occurs when countless decisions flow up the org chart to the very top. Giving business unit leaders full control over key functions allows them to do what is best to meet the needs of their individual units’ customers and maximize their results, and it enables the executives overseeing them to assess their performance. As the Harvard Business School historian Alfred Chandler documented, U.S. companies such as DuPont and General Motors moved from a functional to a multidivisional structure in the early 20th century. By the latter half of the century the vast majority of large corporations had followed suit. Apple proves that this conventional approach is not necessary and that the functional structure may benefit companies facing tremendous technological change and industry upheaval…

Apple relies on a structure that centers on functional expertise. Its fundamental belief is that those with the most expertise and experience in a domain should have decision rights for that domain. This is based on two views: First, Apple competes in markets where the rates of technological change and disruption are high, so it must rely on the judgment and intuition of people with deep knowledge of the technologies responsible for disruption. Long before it can get market feedback and solid market forecasts, the company must make bets about which technologies and designs are likely to succeed in smartphones, computers, and so on. Relying on technical experts rather than general managers increases the odds that those bets will pay off.

Second, Apple’s commitment to offer the best possible products would be undercut if short-term profit and cost targets were the overriding criteria for judging investments and leaders. Significantly, the bonuses of senior R&D executives are based on companywide performance numbers rather than the costs of or revenue from particular products. Thus product decisions are somewhat insulated from short-term financial pressures. The finance team is not involved in the product road map meetings of engineering teams, and engineering teams are not involved in pricing decisions…

It’s easier to get the balance right between an attention to costs and the value added to the user experience when the leaders making decisions are those with deep expertise in their areas rather than general managers being held accountable primarily for meeting numerical targets. Whereas the fundamental principle of a conventional business unit structure is to align accountability and control, the fundamental principle of a functional organization is to align expertise and decision rights.

Ever since Steve Jobs implemented the functional organization, Apple’s managers at every level, from senior vice president on down, have been expected to possess three key leadership characteristics: deep expertise that allows them to meaningfully engage in all the work being done within their individual functions; immersion in the details of those functions; and a willingness to collaboratively debate other functions during collective decision-making. When managers have these attributes, decisions are made in a coordinated fashion by the people most qualified to make them…

The assumption is that it’s easier to train an expert to manage well than to train a manager to be an expert. At Apple, hardware experts manage hardware, software experts software, and so on. (Deviations from this principle are rare.) This approach cascades down all levels of the organization through areas of ever-increasing specialization. Apple’s leaders believe that world-class talent wants to work for and with other world-class talent…

Early on, Steve Jobs came to embrace the idea that managers at Apple should be experts in their area of management. In a 1984 interview he said, “We went through that stage in Apple where we went out and thought, Oh, we’re gonna be a big company, let’s hire professional management. We went out and hired a bunch of professional management. It didn’t work at all….They knew how to manage, but they didn’t know how to do anything. If you’re a great person, why do you want to work for somebody you can’t learn anything from? And you know what’s interesting? You know who the best managers are? They are the great individual contributors who never, ever want to be a manager but decide they have to be…because no one else is going to…do as good a job.”…

Having leaders who are experts in their areas and can go deep into the details has profound implications for how Apple is run. Leaders can push, probe, and “smell” an issue. They know which details are important and where to focus their attention. Many people at Apple see it as liberating, even exhilarating, to work for experts, who provide better guidance and mentoring than a general manager would. Together, all can strive to do the best work of their lives in their chosen area…

Because no function is responsible for a product or a service on its own, cross-functional collaboration is crucial.

When debates reach an impasse, as some inevitably do, higher-level managers weigh in as tiebreakers, including at times the CEO and the senior VPs. To do this at speed with sufficient attention to detail is challenging for even the best of leaders, making it all the more important that the company fill many senior positions from within the ranks of its VPs, who have experience in Apple’s way of operating.

However, given Apple’s size and scope, even the executive team can resolve only a limited number of stalemates. The many horizontal dependencies mean that ineffective peer relationships at the VP and director levels have the potential to undermine not only particular projects but the entire company. Consequently, for people to attain and remain in a leadership position within a function, they must be highly effective collaborators…

In 2006, the year before the iPhone’s launch, the company had some 17,000 employees; by 2019 that number had grown more than eightfold, to 137,000. Meanwhile, the number of VPs approximately doubled, from 50 to 96. The inevitable result is that senior leaders head larger and more diverse teams of experts, meaning more details to oversee and new areas of responsibility that fall outside their core expertise.

In response, many Apple managers over the past five years or so have been evolving the leadership approach described above: experts leading experts, immersion in the details, and collaborative debate. We have codified these adaptions in what we call the discretionary leadershipmodel, which we have incorporated into a new educational program for Apple’s VPs and directors. Its purpose is to address the challenge of getting this leadership approach to drive innovation in all areas of the company, not just product development, at an ever-greater scale…

Whereas Apple’s VPs spend most of their time in the owning and learning boxes, general managers at other companies tend to spend most of their time in the delegating box. Rosner estimates that he spends about 40% of his time on activities he owns (including collaboration with others in a given area), about 30% on learning, about 15% on teaching, and about 15% on delegating. These numbers vary by manager, of course, depending on their business and the needs at a given time…

Apple’s functional organization is rare, if not unique, among very large companies. It flies in the face of prevailing management theory that companies should be reorganized into divisions and business units as they become large. But something vital gets lost in a shift to business units: the alignment of decision rights with expertise.

Apple’s market capitalisation is ~$2.75 trillion — roughly UK annual GDP.

It would be impossible for Whitehall to work like Apple because Whitehall doesn’t believe in expertise, doesn’t train or cultivate or value it. Whitehall actively undermines true expertise with its absurd HR policies. This was stark in the covid disaster when almost every formal meeting involved people talking about X who were not experts on X and almost all useful conversations had to occur away from ministerial meetings.

If Apple worked on Whitehall principles, you’d see the people in charge of the camera suddenly shuffled to be in charge of semiconductors and the people in charge of semiconductors shuffled to be in charge of marketing, all effectively random and then repeated every two years.

I’m not being sarcastic. This is literally how critical Whitehall functions are managed. One minute someone is briefing the PM on special needs education, the next they’re briefing the PM on Russian spy technology. Remember when I wrote about this in January 2020? The media went crazy — but everywhere I went officials said ‘100% true, please try to change it’. Officials know it’s true and the good ones hate it. We started to change it. Everything to do with improving Whitehall changed when we left because the Tories are uninterested in how power works.

Every disaster is the same, a few days of headlines, officials keep their heads down knowing the farce will move on, and before you know it the whole of SW1 is tweeting happily about Russell Brand. Nobody fired, nothing changes, nobody cares, on to the next debacle. Can you even remember the school concrete debacle now?! For ?3 days it was the focus, then gone, nothing changes, in the DfE the guy responsible probably ambled home at 4 unbothered, as they used to back in 2011 when I’d watch the person responsible for the debacle on TV whistle their way out the office in the afternoon, entirely unconcerned.

Now the main priority in Cabinet Office HR, apart from preparing for Labour, is aligning Whitehall with elite universities on race and sex madness.

Also it’s worth noting the incredible lack of ambition among senior civil servants for thinking long-term. Back in 2015 Cabinet Secretary Jeremy Heywood gave a very, very revealing interview to the IFG, an interview I’ve encouraged the covid Inquiry to watch. In it he says many interesting things. Among them, Whitehall has ‘nothing to learn from the private sector’ on project management and ‘digital’. Haha. And he also says regarding ‘horizon scanning’, i.e long-term thinking, that Whitehall can’t be expected to match the resources of … ‘Warwick University’. Res ipsa loquitur…

** On HS2. It was obvious in January 2020 that:

• the project was a total disaster

• Whitehall was chockablock with lies and nonsense trying to justify it including absurd graphs suggesting that demand for HS2 was so high that by mid-century *the entire country* would be either on HS2 or building its successors

• some parts of Whitehall wanted to pull the plug, others were terrified of the whole disaster becoming public in detail and preferred to sweep it all under the carpet at the start of a new parliament

• Boris didn’t care about money — ‘money is just digits in a computer, we can make as many as we want’ as he said many times in 2020.

Here is one of the graphs showing demand rising in a smooth exponential — as you can see in the blue circle, one of the No10 data scientists wrote on it ‘this is bullshit’. This graph was supposed to be shown to the PM to convince him to sign it off. Spads were expected to nod along with it. The Cabinet Office was shocked to find someone in No10 who knew what an exponential curve was and to say in a meeting ‘this graph MUST be wrong’. They withdrew it, blamed DfT and produced more garbage for the PM. The same team was in charge of exponential curves on covid a few weeks later (also an example of what I wrote above: one minute they’re briefing on HS rail, the next on covid).

I tried very hard to cancel it immediately. I pointed out to Boris we were due to spend at least £35 billion just in the 2019-24 Parliament — money that could be truly transformational if spent on high value capital projects, and would be desperately needed if we hit big problems. I pointed out that our ludicrous meetings showed the management of the project by the Department for Transport, HMT and Cabinet Office was catastrophic. I was totally confident (before covid) that this meant the already rapidly increasing budgets would inevitably escalate further and before 2024 everybody would be saying to No10 ‘I’m afraid the numbers have all gone up but we’re in so deep now there’s no alternative PM’. It was obvious that no more money should be poured into such disastrous management. And it was obvious that, as usual, the idea of getting to grips with hopeless management wasn’t on anyone’s to do list — quite the opposite, more billions was, as usual, the alternative to getting serious about management.

I totally failed.

The civil service pushed him to approve. So did clueless Tory MPs and people responsible for the disaster like Osborne. The Chancellor made some vague comments about problems but when the PM made clear he was expected to nod along he piped down and nodded along. The supposed ‘speaking truth to power’ civil service, the people who justify their existence by impartial expert advice with the long-term interests of the country their priority? AWOL. (Like when Boris tried to fix illegal donations and it was the spads who cried foul while the Cabinet Office in 2021-2 tried to help him cover it all up.)

When the PM lay on his possible deathbed in April, some suggested we try to cancel the January decision, given clearly all the numbers would blow up — less demand, more delays, higher costs etc. If he’d died I would certainly have tried hard again to axe it. As it was he returned to work and declared that covid made the case for HS2 ‘even stronger’ and he was ‘even more confident in it’. His argument was that everyone was ‘desperate for sex’ after lockdown so would travel more. Yes, minister… Sums up the ‘business case’ for HS2.

The best time to cancel HS2 was a decade ago. The next best time has always been today. The main reason for my confidence is simple: if the management doesn’t change and the incentives don’t change and we’re left with the normal Whitehall farce, we know how it will work out — the same as it always does. And even though the bills have gone up massively, we can be absolutely confident they will keep going up and up for another 20 plus years. The numbers in HMT already say HS2 will not repay the costs — how much worse will it be in 20 years?

The aircraft carriers were the same. The new ‘optionally crewed fighter’, haha, will be similar. Lies and waste of taxpayers money are endemic. But Insiders will rally around behind all such projects. Starmer will keep ploughing cash in without reform as every PM has post-Thatcher.

HS2 is also a very good example of just how atrocious SW1 generally is at politics, communication and market research.

Osborne sold the Tories on it years ago by telling them it would make them ‘popular in the north’.

'The ‘north’ always thought it’s ‘really shit’.

What did the ‘north’ want? Fast action on intra-north infrastructure.

What did the ‘north’ think about HS2? ‘It’s for London not the north and the corrupt Tories like Osborne are shovelling cash to their mates for land’ etc etc, as focus group after focus group showed.

Yes Insider elements in the north and many who wanted their snouts in the trough obviously supported it.

But politically it was always a disaster.

I told Boris this too in January 2020 when he said ‘George says it’s really popular in the north’. I reminded him how bad Cameron and Osborne were at politics and how our own research showed what they were saying was junk.

Old foxy [his post GE2019 nickname for himself] knows a bit about communication, I will make it sooo popular!

No, you won’t, if you want to be popular make the system dual-carriageway the A1 from Newcastle to Scotland. Obviously, the idea of doing something fast, useful and popular was too radical…

When you watch the pundits and MPs, you see the line everywhere ‘we should have ambition like other countries’. The latter group never bother facing the real question: we can’t do HS2 properly, or other big infrastructure, because of how Whitewall works, if we addressed management failures yes we COULD do infrastructure BETTER than other countries. But nothing, absolutely nothing, not covid, not war, not HS2, not the collapse of the NHS — nothing will persuade political-media-academia elites to discuss management.

---

Bill Gurley: regulatory capture & how Silicon Valley has been so successful ‘because it’s so fucking far away from Washington’

Here is the video of his great and short talk.

Obviously connected to the above: SW1 doesn’t care about regulation and its incredible costs. Academics and think tankers and SW1-twitter claim problems are all ‘the incompetence’ of some hapless minister (unable to exert the slightest control) and a lack of MONEY.

During and after the referendum, the entire Westminster system laughed at the idea that our regulatory system is a nightmare and there are huge gains from changing it.

When we hit covid, this regulatory system directly led to tens of thousands of deaths and hundreds of billions in economic damage, as predicted.

And Westminster stuck resolutely to the script: regulation is irrelevant to why things are broken and expensive and in fact we need more regulation for everything (except trade friction with the EU where regulation is mysteriously super-important).

They’ve stuck to this script through a new war, they’re now literally running out of stuff to send UKR and starting another war, the script will stay the same.

Ever wondered what it was like to live in the Austro-Hungarian Empire pre-1914? Wonder no more, you’re living through similar dynamics…

---

OneWeb, Starlink, LEO satellites

There’s been a lot of coverage recently of Starlink and LEO satellites’ application to war, intelligence etc. Given the Middle East mayhem to come there will be lots more.

Why did I push OneWeb in 2020?

In general things are over-attributed to me. Things that work well that had nothing to do with me, things that are disasters that had nothing to do with me, things that never happened — they got reported as ‘Cummings’ because of how the media works.

Buying OneWeb is an example. It was reported as my idea but it wasn’t.

In summer 2020 an extremely able person I’d hired (who’d worked at DeepMind), Mike Webb, came to me with a plan to take control of OneWeb, the low-earth orbit satellite company. It had gone bankrupt. He’d researched it and spent weeks talking to key players in the global space industry. He said there was a great opportunity. It was exactly the sort of thing we’d gone to No10 to do — create long-term value that the old system couldn’t coordinate to get done.

There was a plan for how to make it work commercially that seemed sound.

But much more important to me were arguments concerning the military, intelligence, and geopolitics. There were ways these satellites could help the conventional military, special forces, and the intelligence services.

China was desperate for us not to grab it so they could grab it and the bandwidth (controlled by international agreements).

The Pentagon was very keen for us to do it partly to block China, partly to help them in various ways. And our defence-intelligence relationship with America relies partly on acquiring niche capabilities that we trade via Five Eyes.

Who thought what?

Ben Wallace and some senior MoD people did not want to do it because they feared they’d be made to pay for it and the MoD budgets were already knackered and had been for years particularly since the disastrous Cameron-Osborne-Heywood spending reviews that broke budgets, broke long-term planning and institutionalised lying in official budgets that would get you jailed in the private sector. (NB. Macpherson, former Permanent Secretary at HMT, very chirpy on twitter about Vote Leave, played an important role in this scandal, a scandal that I guarantee will not be exposed by the next government but buried in more yellow paper.)

Other parts of the MoD and intelligence agencies supported me, including MI6 and GCHQ.

The Cabinet Office was torn between the top of the MOD, other parts of the deep state, and a clear sense that the new No10 was interested in exploring projects like this.

But there wasn’t the machinery to think about such things properly.

This was vividly illustrated when we got the PM, Chancellor, Cabinet Secretary and other senior officials together to discuss it. Boris said:

So who is running this meeting?

And everybody looked at everybody else like the Spiderman meme then laughed.

The British state did not have an institutional mechanism to discuss an issue at the intersection of technology, geopolitics, the MOD, agencies, BEIS, commercial etc.

Fixing this is the sort of thing I spent my time on.

Obviously Boris-Normal-Mode had no interest in the plan but this is where his ‘bad’ features proved handy. My explanation of the actual plan was struggling against his desire to avoid ‘rows’ with some senior officials and boredom about the important issues. I searched desperately for leverage…

Imagine, centuries after Cameron and Osborne are barely footnotes in history, the BORIS SATELLITES will be silently circling the earth, a permanent Augustan monument to YOU…

The eyes glittered… He knew he was being manipulated but:

Verrry innnteresting… mmm [fist crashed down on the table] The People’s Government MUST have our own satellites!

Mission accomplished, the satellites were ours for a bargain price.

To the limited extent Boris was interested in the job of PM, his interest focused on what he described as ‘Augustan monuments’ (to himself) — ‘the hungry sheep look up and must be fed’, he would say, sometimes referring to voters, sometimes to MPs. The only meetings he was really interested in were about bridges, tunnels, bike lanes, trains, buildings. The same bit of his brain that said Yes to every dumb bike lane was the same bit we hacked to buy OneWeb.

The debate now

Obviously we could not explain publicly all the reasons for our decision but you can see from how Starlink has played out in Ukraine some of the obvious applications. E.g It was obvious in 2020 that new possibilities would open by combining the bandwidth from these satellites and drones to improve the speed of OODA loops.

In 2020 Remain-academic Twitter and the SW1 NPCs obviously laughed at us buying OneWeb. Typical Vote Leave they don’t understand anything, haha what use could it be for Britain etc.

The same network predictably responded to leaks about me and others pushing Whitehall to think much more radically about how drones would transform war — haha, how stupid.

The same network now babbles away about StarLink’s value to UKR and the use of drones.

This is the same network who laughed when I wore an OpenAI T-shirt to No10 on my first day to make a point about how technology was going to transform politics and government. Back in 2019, never mind 2016, the consensus was GPT was ‘a toy’.

Back in 2020 the consensus on Putin was much closer to the 2000 consensus on Putin than the 2023 consensus — i.e keep London as ‘the laundry’ for KGB and Putin’s friends to wash their money, let Putin settle scores with impunity, let the parties accept cash from dodgy Russians, ‘London’s lawyers and the City are making lots of cash, don’t rock the boat’ as a senior official summarised it in 2020.

If you look at that Remain-Twitter-NPC network, they have, individually and collectively, no recollection of how everything they said about AI, satellites, drones and Putin just 3 years ago proved to be nonsense. They never recollect, nor are forced to recollect, their ludicrous opinions from a few weeks ago.

CLUNK-CLICK, the NPCs get a software patch and on they go…

How to deal with political ‘news’ as it becomes ever more disconnected from reality?

A. Use Tweetdeck with LISTS for different subjects.

B. Put all political pundits into one list so you can ignore them almost all the time and they don’t pollute using the rest of Twitter. To help, I’ve selected a dozen Twitter accounts highly characteristic of the SW1-swarm and put them on a Twitter list here.

C. (Unless your job specifically demands it…) Stop watching all political news, stop reading any pundits. Stop being one of the people who read ‘news’ as a form of entertainment. It’s depressing and unless you’re very odd it makes you understand important political questions worse.

Instead, once a week max, for 15 mins, scroll through this list. You will see a world of bullshit and delusion. Not only will they share their views on how to manage football teams, ‘what Elon doesn’t understand about running a tech startup’, how weather affects tanks, Putin’s psychology, and all the other things they’re superconfident about — these NPCs will burble about election ‘strategy’, ‘polls’, ‘crucial speeches’ and so on. You’ll see their grasp of politics is as good as their startup management expertise and that the SW1 swarm works by having no memory of what they said a few weeks ago. And there’s clearly something connecting a) them not understanding their day job and b) the self-confidence of people who’ve never built anything and ignore management in all circumstances lecturing the guy who built SpaceX about ‘management’. This personality type dominates politics.

D. Next to this list on Tweetdeck put this other list I’ve made — a list of people building the future. The contrast side by side might inoculate you from wanting to follow mainstream political news. I’ve suggested to a tech company they fine-tune a LLM on SW1’s twitter output to predict how the hacks-MPs-academics will respond to news, I hope they make it public. Technically it’s already possible to automate most political punditry but I assume the old media will sort of continue through 2024, increasingly baffled and decreasingly listened to outside those who went to elite universities. By 2028 though, surely most of these people will have been retired? Big companies are automating ALL their customer service function — all call centres ended. Who will continue paying all these political journalists? Why not have models generate the punditry about the video content generated by other models and cut out the expensive noise-creating middlemen?! More accurate and cheaper!

Also if you watch the SW1 list discuss the ECHR, you’ll see how its biggest defenders will be its biggest liabilities as happened with the EU. Jolyon really wants to be famous. I agree, he should be famous, very famous, his views on the ECHR should be extremely famous — the more famous, the sooner we’ll be out of the ECHR…

A few things to read

Tyler Cowen on Bach.

Scott Alexander wrote about Elon. You’ll see clear evidence for why almost everything you read about Elon is false. Obviously politics-world has no interest in the truth, he’s now — CLUNK-CLICK — a hate figure, a ‘fascist’.

I talked to the BBC for a podcast on campaigns, first episode aired HERE.

Marc Andreessen published a manifesto for techno optimism yesterday. Highly recommended. And Ben Reinhardt’s annotated version!

I just read, but do not recommend you do, the Economist piece — The strange success of the Tories schools policy — purporting to explain ‘why the Tories had so much success on school reform’. It is bad. It repeats every pundit piece on this subject that ‘communication’ was key and a success, attacking ‘the blob’ worked etc. Wrong. Communication was not part of the success, it was a strategic failure with some tactical successes. The ‘blob’ thing, that I’m often (wrongly) credited with, was bad not good (but was popular in the lobby). And it ignores, like every single piece on the subject, the central reason why we got so much done: a) we got rid of ~15/20 of the top people and 4/5 of the very top, b) we totally rebuilt Private Office so we actually controlled the department and c) we built alliances with the younger officials who shared our goals. We also slashed the ‘communication’ team from over 250 to under 50 the day I left and obviously everything improved (and would have improved further if cut to <10). Tory MPs aren’t interested in how we actually gripped a department. And you’re hardly going to hear the truth from people like the Institute for Government. (I also notice one of our former colleagues, Sam Freedman chirps a lot on this subject including at IFG. Back in 2011 he often used to say in the DfE, ‘I know nothing about strategy, management, politics or communication — I’m an expert in school funding’. This was true and honest. But this is very low status. He’s reinvented himself as a pundit on …. strategy, management, politics and communication, with a Blairite political spin. He’s thrown himself so enthusiastically into it he’s joined my favourite pundit trend — the most ineffective people in the world opining confidently about ‘what Elon doesn’t get about managing a tech startup’ — a priceless genre. I admire the chutzpah but he was right in 2011. He is a worthy addition to the NPC-twitter list. The general rule that almost always works is: a) look for the emotional centre of gravity of mainstream political pundits and their assumptions and predictions about how normal voters will respond to communication/politics and b) bet against it being accurate. If you back test this since 2015 it’s almost infallible. Structurally, they have to be behind the curve — if they weren’t they’d be unemployable. And if you watch my NPC Twitter list you will see that these people cannot predict their own use of Twitter, they almost all say, month in month out, ‘it’s intolerable under Elon and I’m off’ — but they don’t leave. You should take their political views as seriously as their predictions about their own social media use…)

---

Fake news and war

The Gaza hospital attack is a perfect example of how the old media is a malignant force that can’t discuss its malignancy.

The prior assumptions of most mainstream journalists are anti-Israel, pro-Palestine.

They have a long history of accepting Hamas statements as if they’re from the Fields Medal Committee.

Last night they immediately reported Hamas’ claim that the hospital was destroyed by an IDF ‘strike’.

The NYT rushed it onto its front page. So did the Guardian, BBC etc.

Then social media started figuring out the facts.

Now videos, geolocation, audio, all sorts of evidence suggests very serious doubt about the original line and I would argue a reasonable person should now assume much less than 50% it was Israel. (Leaving aside complex political inferences like, why would they do this?)

The old media have had to change their headlines.

But of course this happens hours later.

Crowds were already attacking Israeli Embassies.

After the Hamas attacks the universities were full of ‘show us the evidence we can’t trust Israel’. But when Hamas puts something out the prior is to believe it — Israel = fascist warmongers!

Obviously you won’t see this discussed properly in the old media.

This is all part of how technology has fragmented consensus reality.

And it can only accelerate with the spread and growing power of generative models like GPT.

Another irony of the current situation. Our universities are full of academics who believe the dopey conspiracy that a KGB operation pulled off Trump and Brexit using Facebook. The same academics overwhelmingly have views on Israel shaped by an actual multi-decade massive KGB propaganda operation since the 1960s, now extensively documented.

There are MANY academics who agree with this Cornell Professor who found the experience of watching the Hamas massacre ‘exhilarating’, but probably don’t know WHY they find it so.

Given all the things we’ve done to kill Russians at scale, what are the odds that Putin helps someone in the Middle East do something like sink a US ship, blow up a US Embassy or base etc? And Russia just says, well, run around the world like NATO and bad shit happens. I’m not talking in moral terms — who is right/wrong. I’m just talking about probabilities given how key players think/feel. As I’ve pointed out for years, very expensive ships are much more vulnerable to relatively cheap missiles than ever and innovations driven by the UKR war are accelerating these trends. This is why in war games the UK carriers flee to the edge of the game at the start…

---

I read (much not all of) this recently as part of my Bismarck project, it’s very good even if arguably a bit too pro-Metternich)…

---

Thanks for subscribing.

More imminently…