Artificial Intelligence: The Control Dichotomy

Professor Nick Bostrom delivered an inspirational speech for the SCL last night on "Superintelligence: a godsend or a doomsday device", although few would have found it reassuring - it is certainly conceivable that machines could become more intelligent than humans and that humans might not be able to control them. But these are still early days, he stresses. Regulating the development of artificial intelligence at this point risks halting progress. There's a lot more work to do to really understand how artificial intelligence will develop beyond playing old video games better than humans or recognising an image as the picture of a cat. We need to consider how the technology could help avert our extinction, as well as how it might wipe us out. Yet Nick also warns that it will take far less time for computers to exceed human level machine intelligence than to reach it in the first place. So we need to start work on the control mechanisms for the development and use of artifical intelligence, without regulating the industry out of existence: the control dichotomy.

Nick suggests that the guiding principle should be that of "differential technological development" - diverting resources away from technologies and their application which could cause human extinction while focusing on those which will either help prevent our demise or will facilitate the expansion of the human race throughout the cosmos.

But how do we distinguish between helpful and harmful technologies and their application? 

As Nick points out, it's tough to think of any human invention that is inherently 'good'. He mentions many things, from gun powder to genetic engineering, and I think we can throw in the wheel and the Internet for good measure. All these things are used by humans in bad ways as well as for the greater good. But I think what horrifies us especially about the idea of Superintelligence or 'The Singularity' is that it will be machines, not bad humans, who will be using other machines against us. And while we have lots of experience in dealing with evil humans, even our top minds admit we still don't know much about how machines might act in this way or how to stop them - and what we humans fear most is the unknown. 

You'll notice I haven't said 'evil' machines, since they might not be operating with any evil 'intent' at all. Human extinction might just be a mistake - 'collateral damage' arising from some other mission. For instance, Nick suggests that a particular machine left to itself in a given situation might decide to devote itself entirely to making paperclips. So, presumably, it would not bother to put out a fire, for example, or free a human (or itself) from a burning building. It might leave that to other machines, who might in turn have chosen their own narrow objective that involves ignoring people's screams.

Here's where I struggle with the notion of Superintelligence. In fact, as someone who hates being pigeon-holed into any single role, I think a machine's decision to only ever make paperclips might be fabulously logical and a brilliant choice for that machine in the circumstances, but it makes the machine as dumb as a post. For me, Superintelligence should involve a machine being able to do everything a human can and more. 

But that's beside the point. Knowing what we know already, it would be insane to ignore the paperclip droid and wait for artificial intelligence to develop a machine more capable than humans before figuring out how we might control it. Nick is right to point out that we must figure that out in parallel. In other words, the concept of human control has to be part of the artificial intelligence programme. But it won't be as simple as coding machines to behave protectively, since machines will be able to programme each other. For instance, Nick suggests we could put the machines to work on the control problem, as well as on the problem of how to ensure the survival of our species. AI labs might also pay insurance premiums to cover the damage caused by what they develop. He was less certain about what we might do to constrain developments that occur in the context of secret defence programmes or intelligence gathering, but he seemed confident that we could at least infer the pace of development from the results, and be able to consider how to control the wider application of those developments. Mmmm.

At any rate, Nick also warns that we need to be careful what we wish for. Mandating human survival in a prescriptive way - even a specific biological form - would be a bad move, since we should not assume we are in a position to foster positive human development any more than the Holy Office of the Spanish Inquisition. Better to embed positive human values and emotions or, say, entertainment as a feature of intelligent machines (although I'm guessing that might not go down well with the jihadis). From a phyiscal standpoint, we already know that the human body won't do so well for long periods in space or on Mars, so some other version might need to evolve (okay, now I'm freaking myself out).

To retain a sense of pragmatism, at the end of the speech I asked Nick what he would recommend for our focus on 'Keeping Humans at the Heart of Technology' at the SCL conference in June. His tip was to consider which of the various types of control mechanism might work best, recognising the need to avoid constraining the positive development of artificial intelligence, while ensuring that we will be able to keep the machines in check if and when they become smarter than us.

No pressure then...

Will Machines Out-Compete Humans To The Point of Extinction?

I've been a bit absent from these pages of late, partly pulling together SCL's Technology Law Futures Conference in June on 'how to keep humans at the heart of technology'. As I've explained on the SCL site, the conference is part of SCL's effort to focus attention on that question all year, starting with a speech by Oxford University's Professor Nick Bostrom on 2 March: "Superintelligence: a godsend or doomsday device? 

In other words, last year was when the threat of "The Singularity" really broke into the mainstream, while this year we are trying to shift the focus onto how we avert that outcome in practical terms. 

My own book on how we can achieve control over our own data is still ping-ponging between agent and publishers, but will hopefully find a home before another year is out - unless, of course, the machines have other ideas... 

Humans, Not Big Data, Must Benefit From Opening Up The Banks

One of the 'good news' items in last year's Autumn Statement was the endorsement of an Open Data Institute report on how innovators could make better use of bank data on our behalf (instead of on the banks' behalf). Last week, the Treasury followed up with a call for evidence "on how best to deliver an open standard for application programming interfaces (APIs) in UK banking and... whether more open data in banking could benefit consumers."

This is important, because enabling our own machines to start crunching our financial data on our behalf is key to humans winning the race against Big Data.

'Open data' involves enabling public access to data by connecting data sets using uniform resource identifiers ('Linked Data') and publishing data about them in machine-readable formats. While the government and other public institutions have done a good job of opening up public data sets so far, our private institutions are lagging behind, as discussed here. That's because they benefit from making sure we know less than them about our own requirements - 'information asymmetry'.

The Midata initiative was the first to aim at restoring the balance in favour of consumers, by taking aim at energy utilities, telecoms providers and banks. Now the energy companies face being opened up like so many tin cans by smart metering, while the telecoms market looks like it will go from bad to worse. Banks have also proved a tougher nut to crack. Yes, some people can download their current account data via internet banking. But in addition to consumers, there are 4.5 million small businesses out there that need a hell of a lot more than that. So far, it's taken regulatory action just to start the process of improving access to small business credit data and making a market in rejected small business loan applications. Now the Treasury is trying to force the pace on the technology necessary to support all that.

The ODI's recommendations for opening up banking are: 

• banks should agree an open API standard to support third party access to bank data - basically firms that can help you make sense of the data (but Big Data firms will try to persuade you to share it with them too);
• independent guidance should be provided on technology, security and data protection standards that banks can adopt to ensure data sharing meets all legal requirements; 
• an industry wide approach should be established to vet third party software applications and publish a list of vetted applications as open data - this would allow visibility of firms that are acting on consumers' behalf and those who are not;
• standard data on Personal Current Account terms and conditions should be published by banks as open data; and 
• credit data should be made available as open data.

My main concern is that requiring agreement on 'standards' as a precondition for opening up banking will enable the banks to delay the whole process by a decade - as they did with Faster Payments. As was recommended by the security working group in the Midata initiative, I would prefer to see banks required to immediately make their data available to each consumer/SME in whatever open format the banks choose, while adhering to common data security protocols, and leave it to the open data community to figure out how to re-format and display whatever rubbish they dish up.

Apart from complaints by the banks, Treasury officials expect to hear positive contributions from consumer groups, other financial services providers, financial technology firms and app and software designers.

Let's not disappoint them. This is a good opportunity to ensure the government clears the way for innovation that puts you and me at the heart of financial services, without mistakenly creating further barriers in the process.

The P2P Jobs Market

The UK has an army of 4.6 million self-employed people, according to the Office of National Statistics - the largest it's been since we began recording such figures 40 years ago. That's 15% of the UK workforce. Even more significant is that 732,000 of the 1.1m people who have found work since early 2008 are self-employed. Fewer people have been leaving self-employment for employee roles over the past five years than used to be the case. Perhaps that's because of the recession. But over that period numerous services have emerged to support self-employment, and it seems possible that we'll see an even greater shift towards that way of working in the future.

So, who are the self-employed, and how do they find work?

The largest increase in self-employment since 2008 has been among 'managers, directors and senior officials'. But, hey, every self-employed person might claim to be a manager. So it's more noteworthy that the top 3 self-employed roles of 2014 have been building trades, cab drivers/chauffeurs and carpenters. The figures also show that professional and technical occupations are heavily represented.

How do these people find work?

No doubt word of mouth has a lot to do with it. But we've also seen a rise in the number of online marketplaces that match self-employed people with those who need work done. Indeed, TaskRabbit, a leading US marketplace, chose London as testing ground for a more automated model that it later rolled out in the US to replace its initial manual auction service. While TaskRabbit currently seems to cover the broadest range of services, there are many other such marketplaces in the UK, such as RatedPeople, Trustatrader, MyHammer, MyBuilder, TradeAdvisor, Checkatrade and so on. Note that Amazon has launched a 'local services' offering in the US, which suggests it may one day do so here.

The prevalence of cab drivers amongst the self-employed may help explain the growing number of taxi apps and car-share services.

Meanwhile, SchoolofEverything (a client of mine from 2007, on the back of my experience of P2P lending), enables anyone to make money from giving lessons in almost anything you could think of (...no, not that).

At any rate, the growth in both the number of self-employed people and the services that help them find work, suggests that self-employment could be an even more popular model in the future. The rise of the P2P economy?

Another Hung Parliament, Please

With the UK general election looming in May, I thought I'd declare my apolitical hand: I'm hoping for another 'hung' Parliament and a coalition government.

I've been a fan of the idea since the opportunity presented itself in the last general election. I think the beast has worked pretty well for the pragmatic amongst us, and is well suited to dealing with the nasty challenges ahead. As I hoped in April 2010, politicians on both sides of the coalition have had to behave much more reasonably and responsibly in seeking solutions to the root causes of our problems than their party-political dogma would have otherwise dictated. This has spiked the guns of an extremely dogmatic opposition. And even the media's doom-mongering about instability and chaos has proved groundless. Sure there have been U-turns and major disagreements between the coalition parties, but the democratic progress should be dynamic, open and messy - not engineered, top-down, by a party leader with a Whip.

The same form of government is needed over the next five years because the long journey out of the tunnel has barely begun. That light up ahead is not looming economic recovery, it's an on-coming train laden with vast public sector debt, slowing Chinese growth, savagely low oil prices that might rebound higher than before, the Russian Problem, insanity in Greece, negative real interest rates and a stagnant Eurozone. Oh, and a new global financial crisis, as Hank Paulson infamously forecast in 2010:

"...We'll have another financial crisis sometime in the next 10 years because we always do.""

The public finances are still in a parlous state. So all the UK political parties face the need to cut public spending, whether they like it or not. Raising expenditure is out of the question, because it would mean borrowing more - and higher taxes won't bring in any more money. The total UK tax receipts have hovered at or below 40% of GDP for over 40 years. We're bumping along the ceiling, people! Raise taxes, the economy grinds to a halt and the best you'll get is 40% of a smaller pie. Cut taxes to around 35% of GDP,  the economy roars into life and you get a smaller slice, but of a much bigger pie.

But, hey, if you think the UK should drift into the next financial crisis with even higher debt and taxes, why not simply move to Greece?

What's left to cut? There's no end to it: we need our politicians and civil servants to remain focused on making the public sector more efficient, by removing waste and insisting that services be designed to operate more efficiently in future, particularly in the major spending areas. The defence budget, for example, is a rounding error on a more efficient tax and benefits system and a leaner, better co-ordinated public health and social care sector (20% of hospital beds are occupied by people who aren't even sick!). Money could also be saved by addressing root causes instead of their many symptoms. For instance, would more social housing have helped ease the pressure on first-time buyers, avoiding government subsidies to them and pre-empted the policy battle over immigration levels? Similarly, we must continue financial reforms to increase the sources of funding and the range of payment services for consumers and small businesses (who create half of all new jobs) because the economy is still too dependent on a few major banking groups who remain a millstone around the country's neck.

Some people will say this is dry, boring and unimaginative. But if you want entertainment, head to the movies. 

Others want something to believe in. For instance, they accuse David Cameron of lacking political ideology or a 'pattern of belief', an '-ism'. Yet they claim that his "legacy will be a collection of tactical manoeuvres, with as many prominent surrenders as victories." Apparently these people have never heard of pragmatism. But they've also unwittingly hit on the benefit of the hung Parliament in restraining coalition parties from implementing their more extreme policies. By contrast, the 'believers' expect us to cling to the idea that Ed Miliband is "in politics for the right reason" (just the one?) or "propelled by something more noble than the salvation of his own skin", which you could choose to mean anything that gets you through the day. But beware words like 'right' and 'noble'. They are the cloaks of dogma and moral panic - rallying cries for the likes of Tony Bliar's weird crusade or Gordo's crash, in which Miliband (and Balls) played key roles - not to mention the ballooning cost of the Security State. So, actually, if we believe anything in this vein, then it's surely that such 'noble' ambitions make Labour governments the kind of luxury that only a much wealthier country could afford. 

But, who knows, maybe being trapped in a coalition would even convert Ed to pragmatism.

Whichever way you look at it, we need a government that's forced to focus on resolving the root causes of society's actual problems - not one driven to distort the facts to suit its own dogmatic solutions. And my sense is that only another hung Parliament will ensure we get it.