![](\"https:\/\/tctechcrunch2011.files.wordpress.com\/2016\/09\/gettyimages-498915290.jpg?w=738\")
<\/p>\n<\/p>\n
What are tech titans Google, Amazon and Uber agitating\u00a0for to further the march\u00a0of machine learning technology and ultimately inject more fuel in the engines of\u00a0their own dominant platforms? Unsurprisingly they\u2019re after access to data. Lots and lots of\u00a0data.<\/p>\n
Specifically,\u00a0they\u2019re pushing for free and liberal access\u00a0to\u00a0publicly funded data\u00a0\u2014 urging\u00a0that this type of\u00a0data continue to be \u201copen by default\u201d, and\u00a0structured in a way that supports \u201cwider use of research data\u201d. After all, why pay to acquire data when there are vast troves of publicly-funded information\u00a0ripe\u00a0to\u00a0be squeezed for fresh economic gain?<\/p>\n
Other items\u00a0on this\u00a0machine learning advancement wish-list include new open standards for data (including metadata); research study design that has\u00a0the \u201cbroadest consents that are ethically possible\u201d; and a stated desire to rethink\u00a0the notion of \u2018consent\u2019 as a core plank\u00a0of good data governance \u2014\u00a0to grease the pipe in favor of\u00a0data access and make data holdings\u00a0\u2018fit for purpose\u2019 in\u00a0the AI\u00a0age.<\/p>\n
These suggestions\u00a0come in a 125-page report<\/a> published today by the Royal Society, aka the UK\u2019s national academy of science, ostensibly aimed at fostering\u00a0an environment where\u00a0machine learning technology can flourish in order to unlock mooted productivity gains and economic benefits \u2014 albeit the question of who, ultimately, benefits as more and more data gets squeezed to give up\u00a0its precious\u00a0insights is the overarching theme and unanswered\u00a0question here. (Though the supportive presence of voices\u00a0from three of tech\u2019s most powerful machine learning deploying platform giants suggests one\u00a0answer.)<\/p>\n The report, entitled Machine learning: the power and promise of computers that learn by example<\/em>, is\u00a0the work of\u00a0the Royal Society\u2019s working group on machine learning<\/a>, whose\u00a015-strong membership includes employees of three companies currently deploying machine learning at scale:\u00a0Demis Hassabis, the founder and CEO of Google DeepMind, along with DeepMind\u2019s research scientist, Yee Whye Teh; Neil Lawrence, Amazon\u2019s director of machine learning; and Zoubin Ghahramani, chief scientist at Uber.<\/p>\n The report\u2019s top-line\u00a0recommendations boil down the\u00a0more fleshed out\u00a0concerns in the meat of its\u00a0chapters, and end up foregrounding encouragement at greater length than concern, as you might expect from\u00a0a science academy \u2014 though the level of concern contained within\u00a0its pages is notable nonetheless.<\/p>\n The report recommendations\u00a0laud\u00a0what is\u00a0described as\u00a0the UK\u2019s\u00a0\u201cgood progress\u201d in increasing the accessibility of public sector data, urging \u201ccontinued efforts\u201d towards\u00a0\u201ca new wave of \u2018open data for machine learning\u2019 by government to enhance the availability and usability of public sector data\u201d, and calling for\u00a0the government to \u201cexplore ways of catalysing the safe and rapid delivery\u201d of new open standards for data which \u201creflect the needs of machine-driven analytical approaches\u201d.<\/p>\n But\u00a0an early glancing reference to\u00a0\u201cthe value of strategic datasets\u201d does get\u00a0unpacked in more detail further into the report\u00a0\u2014 with the\u00a0recognition that early access to such valuable troves of publicly-funded data could\u00a0lock in commercial advantage. (Though you won\u2019t find a single use of the word \u2018monopoly\u2019 in the entire document.)<\/p>\n \u201cIt is necessary to recognise the value of some public sector data. While making such data open can bring benefits, considering how those benefits are distributed is important,\u201d they write. \u201cAs machine learning becomes a more significant force, the ability to access data becomes more important, and those with access can attain a \u2018first mover feedback\u2019 advantage that can be significant. When there is such value at stake, it will be increasingly necessary to manage significant datasets or data sources strategically.\u201d<\/p>\n There is no\u00a0example of this kind of \u201cfirst mover feedback advantage\u201d set out\u00a0in the report, but you could point to DeepMind\u2019s data access partnerships<\/a>\u00a0with the UK\u2019s National Health Service as a pertinent case study here. Not least as the original data-sharing arrangement that the Google-owned company\u00a0reached with the Royal Free NHS Trust in London is\u00a0controversial<\/a>, having been agreed without patient knowledge or consent, and having scaled significantly in scope from its launch\u00a0as a starter app hosting an NHS algorithm to (now) an ambitious\u00a0plan to build a patient data API<\/a>\u00a0to broker third party\u00a0app makers\u2019\u00a0access to NHS data. Also relevant but unmentioned: the original DeepMind-Royal Free data-sharing agreement remains under investigation by UK data protection watchdogs<\/a>.<\/p>\n Instead, the report flags\u00a0up\u00a0the value of NHS data \u2014 describing it as \u201cone of the UK\u2019s key data assets\u201d \u2014 before going on to frame the notion of third party\u00a0access to UK citizens\u2019\u00a0medical records as a case of \u2018personal privacy vs\u00a0public good\u2019, suggesting that \u201cappropriately controlled access mechanisms\u201d could\u00a0be developed to resolve what it dubs this \u201cbalancing act\u201d (again, doing so without mentioning that\u00a0DeepMind has already set itself the self-appointed task of developing a controlled access mechanism<\/a>).<\/p>\n \u201cIf this balancing act is resolved, and if appropriately controlled access mechanisms can be developed, then there is huge potential for NHS data to be used in ways that will both improve the functioning of the NHS and improve healthcare delivery,\u201d they write.<\/p>\n Yet exactly who stands to benefit economically from unlocking valuable healthcare insights from a\u00a0publicly funded\u00a0NHS is not discussed. Though common sense would tell you that\u00a0Google\/DeepMind believes there is a\u00a0profitable business to be built off of free access to\u00a0millions of\u00a0NHS patient\u2019s health data and the first mover advantage that gives them \u2014 including the chance to embed themselves into healthcare service delivery via control of an access infrastructure.<\/p>\n In an accompanying\u00a0summary to\u00a0the\u00a0report, a pull quote from another member of the working group, Hermann Hauser, co-founder of Amadeus Capital Partners, talks excitedly about potential transformative opportunities for businesses\u00a0making use of machine learning\u00a0tech. \u201cThere are exciting opportunities for machine learning in business, and it will be an important tool to help organisations make use of their \u2014 and other \u2014 data,\u201d he is quoted as saying. \u201cTo achieve these potentially significant economic benefits, businesses will need to be able to access the right skills at different levels.\u201d<\/p>\n The phrase \u201ceconomic benefits\u201d is at least mentioned here. But the raison d\u2019etre<\/em> of investors\u00a0is to achieve\u00a0a good exit. And there has been a rash of exits of machine learning firms to big tech giants engaged in the war for AI\u00a0talent. DeepMind selling to Google for more than $500 million<\/a>\u00a0in 2014 being just one example.\u00a0So\u00a0investors have their\u00a0own dog in the\u00a0fight\u00a0for a less stringent public sector data governance regime \u2014 and still get to cash out if an AI startup they bet on sells to a tech giant, rather than scales\u00a0into one itself.<\/p>\n Julia Powles, a tech law and policy researcher at Cornell Tech makes short shrift of the\u00a0notion\u00a0that lots of entrepreneurs stand to benefit if the public sector data floodgates are opened.\u00a0\u201cThe idea that small guys can make use of their data is just a ruse. It\u2019s only the big that will profit,\u201d she tells TechCrunch.<\/p>\n Another portion of\u00a0the report spends a lot of time\u00a0apparently concerned with skills \u2014 discussing ways the\u00a0government could\u00a0encourage \u201ca strong pipeline of practitioners in machine learning\u201d, as it puts it \u2014 including urging it to make\u00a0machine learning a priority area for additional PhD places, and to make near-term funding available for 1,000 extra PhDs (or more). Machine learning PhDs are of course top of the hiring tree\u00a0for big tech giants who have the most cash to suck up these\u00a0highly prized\u00a0recruits, keeping them from being hired by startups, or indeed from starting their own competing businesses. So any increase at the top academic tier will be\u00a0Google et al\u2019s gain, first and foremost \u2014 more so if the public sector also paid to fund these extra PhD places.<\/p>\n The skills discussion (which includes suggestions to tweak school curriculum to include\u00a0machine learning over the next five years) has to\u00a0later be weighed against another portion\u00a0of the report considering the potential impact of AI\u00a0on jobs. Here the report cannot avoid the conclusion that machine learning will at very least \u201cchange\u201d work \u2014\u00a0and may well lead to seismic shifts in the employment prospects for large swathes of the workforce which could also, the authors recognize, increase\u00a0societal inequality. All of which does rather undermine the earlier\u00a0suggestion that \u2018everyone\u2019 in the society will be able to\u00a0upskill for a machine learning driven\u00a0future, given you can\u2019t acquire\u00a0skills\u00a0for jobs that don\u2019t exist\u2026 So the risk of AI generating a drastically asymmetric wealth and employment\u00a0outcome is both\u00a0firmly\u00a0lodged in the report\u2019s vision of future work \u2014 yet also\u00a0kicked into a\u00a0no-man\u2019s-land\u00a0of collective (i.e. zero ownership) responsibility.<\/p>\n \u201cThe potential benefits accruing from machine learning and their possibly significant consequences for employment need active management,\u201d they write. \u201cWithout such stewardship, there is a risk that the benefits of machine learning may accrue to a small number of people, with others left behind, or otherwise disadvantaged by changes to society.<\/p>\n \u201cWhile it is not yet clear how potential changes to the world of work might look, active consideration is needed now about how society can ensure that the increased use of machine learning is not accompanied by increased inequality and increased disaffection amongst certain groups. Thinking about how the benefits of machine learning can be shared by all is a key challenge for all of society.\u201d<\/p>\n Ultimately, the report does call for\u00a0\u201curgent consideration\u201d to be given to what it describes as \u201cthe \u2018careful stewardship\u2019 needed over the next ten years to ensure that the dividends from machine learning\u2026 benefit all in UK society\u201d. And it\u2019s true to say, as we\u2019ve said before<\/a>, that policymakers and regulators do need to step up and start building\u00a0frameworks and determining rules to ensure machine learning technologists do not have the chance to asset strip the public sector\u2019s crown jewels before they\u2019ve even been valued (not to mention leave\u00a0future citizens unable to pay for the fancy services that\u00a0will then be sold back to them, powered by machine learning models\u00a0freely fatted up on publicly funded data).<\/p>\n But the suggested ten year timeframe seems disingenuous, to put it mildly. With \u2014 for instance \u2014 very large quantities\u00a0of sensitive NHS data already flowing from the public sector into the hands of one of the world\u2019s most market capitalized companies (Alphabet\/Google\/DeepMind) there would seem\u00a0to be rather more short term urgency for policymakers to address this issue. Not leave it on the backburner for a decade or so. Indeed, parliamentarians have already been urging action on AI-related concerns\u00a0like algorithmic accountability<\/a>.<\/p>\n Public opinion is understandably\u00a0a big preoccupation for the report authors \u2014 unsurprisingly so, given that\u00a0a technology that potentially erodes\u00a0people\u2019s privacy\u00a0and<\/em>\u00a0impacts their\u00a0jobs risks being\u00a0drastically unpopular. The Royal Society conducted\u00a0a public poll on machine learning for the report, and say they found mixed views among Brits. Concerns\u00a0apparently included \u201cdepersonalisation, or machine learning systems replacing valued human experiences; the potential impact of machine learning on employment; the potential for machine learning systems to cause harm, for example accidents in autonomous vehicles; and machine learning systems restricting choice, such as when directing consumers to specific products and services\u201d.<\/p>\n \u201cOngoing public confidence will be central to realising the benefits that machine learning promises, and continued engagement between machine learning researchers and practitioners and the public will be important as the field develops,\u201d they add.<\/p>\n The report suggests that large-scale machine learning research programs should include funding for \u201cpublic engagement activities\u201d. So there may at least, in the short term, be jobs for PR\/marketing types to put a good spin on the \u2018societal benefits of automation\u2019. They also call for\u00a0ethics to be taught as part of postgraduate study so that\u00a0machine learning researchers are given \u201cstrong grounding in the broader societal implications of their work\u201d. Which is a timely\u00a0reminder that most of the machine learning tech already deployed in the wild, including commercially, has probably been engineered and implemented\u00a0by minds lacking such a strong ethical grounding. (Not that we really need reminding<\/a>.)<\/p>\n \u201cSociety needs to give urgent consideration to the ways in which the benefits from machine learning can be shared across society,\u201d\u00a0the report concludes. Which is another way of saying\u00a0that machine learning risks concentrating wealth and power in the hands of a tiny number of massively powerful companies and individuals \u2014 at society\u2019s expense. Whichever way you put it, there\u2019s plenty of food for thought here.<\/p>\n Featured Image: elenabs\/Getty Images<\/small><\/p>\n","protected":false},"excerpt":{"rendered":"Scramble for public data\u00a0<\/strong><\/h3>\n
Seismic shifts<\/strong><\/h3>\n
Perception and ethics<\/strong><\/h3>\n
![\"\"](\"https:\/\/tctechcrunch2011.files.wordpress.com\/2016\/09\/gettyimages-498915290.jpg?w=680\")
What are tech titans Google, Amazon and Uber agitating for to further the march of machine learning technology and ultimately inject more fuel in the engines of their own dominant platforms? Unsurprisingly they’re after access to data. Lots and lots of data. Read More<\/a><\/p>\n\n<\/img><\/a>
<\/img><\/a>
<\/img><\/a>
<\/img><\/a>
<\/img><\/a>
<\/img><\/a>\n<\/div>\n
<\/p>\n","protected":false},"author":5817,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","footnotes":""},"categories":[25],"tags":[59,66,26],"class_list":["post-339181","post","type-post","status-publish","format-standard","hentry","category-technology","tag-media","tag-techcrunch","tag-technology"],"_links":{"self":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts\/339181","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/users\/5817"}],"replies":[{"embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/comments?post=339181"}],"version-history":[{"count":0,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts\/339181\/revisions"}],"wp:attachment":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/media?parent=339181"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/categories?post=339181"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/tags?post=339181"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}