Over the course of this week two government reports on the role of artificial intelligence have been launched, officially confirming the arrival of the technology on a global scale.
The tech industry has been all over AI for a while but, so far, governments have stayed quiet. With the launch of the Science and Technology Committee’s report on Robotics and Artificial Intelligence in the UK and the White House’s National Science and Technology Council Committee on Technology report, AI has now been thrust into central conversations at a regulatory and policy level.
The tone of both the reports is largely what you would expect. The UK report, which was prepared by Science and Technology Committee, was pessimistic on the country’s current readiness for AI, while also critical on government actions to date. The US report was largely negative in its tone when addressing the future, focusing on correcting the challenges than growing the potential, and at times demonstrated paranoia more than anything else. In both reports, there was a very strong undertone of haughty self-importance.
But what did we learn from the reports?
Education and Skills
Both reports focused on one area which has been clear for some time, though has not been recognized in an official capacity: the education systems in both the UK and the US are not built towards creating the skills required for a digital era and artificial intelligence. The industry is already in a skills drought, this can be seen through the vast sums of money which is being spent on AI start-up acqui-hires, and it will only become worse.
Within the UK, the Science and Technology Committee has called upon the government to stop dilly-dallying and commit to a digital strategy which would trickle down to the education system. Professor Rose Luckin from the UCL Institute of Education commented “I do not feel that the moment we are equipping either students in school or workers in the workforce with the requisite skills to know how to adapt themselves to use the automation they are being offered to the best effect.” In short, Rose believes the education system needs to evolve to prepare students for a world where AI and robotics are commonplace.
The US government also echoes this point. “Integrating AI, data science, and related fields throughout the Nation’s education system is essential to developing a workforce that can address national priorities. Educational institutions are establishing and growing AI programs at all levels. Universities, colleges, and even secondary schools are expanding AI and data science curricula, but more programs and teachers are needed.”
One of the most important notes from both the reports is a change in mind set to both individuals and the education system. While the millennials are accustomed to the connected world, the education system is not. Schools and universities need to adapt to create a more data-literate community to alight with the digitally-orientated economy which is on the verge of mainstream.
Economy and jobs
This is an area which Telecoms.com has discussed before, and has been addressed by a number of industry commentators but not with any official substance to date. The introduction of artificial intelligence will have to managed responsibly to ensure the impact on employment in certain sectors is not too drastic.
The UK report has taken a very balanced, if non-committal view on the topic, highlighting the number of jobs which will be created alongside the ones which will be lost. While this is encouraging for the country and economy on the whole, your correspondent isn’t too sure the same sentiment will be felt by a London cabby having to retrain in his mid-fifties.
The White House’s National Science and Technology Council Committee on Technology instead raise an interesting point which has so far flown under the radar.
“Analysis by the White House Council of Economic Advisors (CEA) suggests that the negative effect of automation will be greatest on lower-wage jobs, and that there is a risk that AI-driven automation will increase the wage gap between less-educated and more-educated workers, potentially increasing economic inequality.”
In a country where tensions are already high, with the difference between the increasingly wealthy middle-class and the further down-trodden working classes plainly obvious, the transition will have to be managed carefully to ensure the negative impacts are not too drastic. The roll-out of artificial intelligence will come faster than any other scientific advancement in history, how the government reacts and the policies created to support redundant workers could partially define the success of AI.
A question of ethics
How do you programme ethics and morals into a computer? If philosophers have been pondering the questions of existence, morality and consciousness for millennia without coming to a conclusion, what chance do plebs like us have of translating that unknown answer into a language a computer can understand.
AI will impact every aspect of day-to-day life from self-driving cars to medical diagnosis, to intelligent tutoring programmes, raising a host of social, legal and ethical questions. Transparency in the decision making process of the machine has to be paramount, to maintain the AI is continuing to act within the parameters set forward. But who writes these parameters in the first place? Preventing bias being built into the AI system could be a very complicated task.
The report from the White House has even gone as far as to suggest introducing new training programmes for those involved in the development of AI. “Ethical training for AI practitioners and students is a necessary part of the solution. Ideally, every student learning AI, computer science, or data science would be exposed to curriculum and discussion on related ethics and security topics. However, ethics alone is not sufficient.”
The question of ethics and control models in artificial intelligence could be a melting pot of political, economic and social conflicts in the future. International collaboration is key though the one-upmanship which is commonplace in the political world and standards development could see a slight delay.
Both reports demonstrated a clear message: we’re the best and should lead the development of AI on a global scale. The ‘listen to me’ attitude should hardly come as a surprise considering politicians and academics are some of the more self-righteous career paths available, though there is certainly a bit of international posturing coming through too.
The report has highlighted the UK considers itself the self-appointed world leader in artificial intelligence, and thus the UK government should play a central role in the development of international standards. Back in 2013, the UK government identified Robotics and Artificial Intelligence as one of the ‘Eight Great Technologies’ which should be supported and cultured. A committee was promised and to-date, three years later, has not appeared. This contradictory chain of events dents the UK government’s claim for prominence in the international AI arena.
The US government is hardly any better. Once again throughout the report there are several undertones of ‘we’re going to be the boss at this’. The posturing is hardly surprising as the potential of AI is immense; governments will want international policy to best support its own citizens needs and businesses ambitions.
There isn’t necessarily a right answer to which country is most advanced at the moment. The US has Silicon Valley, and some of the leading minds in the technology world. The UK has Google’s Deepmind headquartered in London, a group of machine learning engineers who are arguably the best in the world. The Germans meanwhile, who are currently hosting the global IBM Watson HQ, have remained uncharacteristically quite over this international land-grab.
Pompous public sector
Another area which will have to be kept an eye on is the role of government. Throughout both reports there were numerous claims of the importance of the government in the development of the industry. Yes, there is a role from a regulatory and policy perspective, as well as education, but the involvement in the development of the technology should be managed because, let’s be honest, the best engineers are going to be working for the AI companies not the government.
In short, the role of the government needs to be defined. After reading both reports it was still unclear how involved the government needs or wants to be. For example, there are several statements made in the reports which suggest the development of AI couldn’t possibly go any further without government assistance. Somehow Google got this far without public servants helping out whenever they decide to turn up to the office.
“Without a Government strategy for the sector, the productivity gains that could be achieved through greater uptake of the technologies across the UK will remain unrealized” – UK report
“Provide much need direction and coordination” – UK report
“Given the strategic importance of AI, moreover, it is appropriate for Federal Government to monitor developments in the first worldwide” – US report
“Recommendation 12: Industry should work with government to keep government updated on the gradual progress of AI in the industry, including the likelihood of milestones being reached soon” – US report
The government is important in the development of new technology, though it should be put in context. The private sector has more R&D funds, better insight on the practical implementations of AI and more qualified engineers to move the technology forward. Government has an important role to play, but let’s keep it to the right areas.
One quote which should be analysed further is in the UK report in the section focused around standards and regulation. “It should focus [a committee to oversee the development of AI] on establishing principles to govern the development and applications of AI techniques, as well as advising government of any regulation required on limits to its progression.”
Limits are needed, no-one wants AI to take over the world, but the limits imposed should be closely scrutinized to ensure progress is not faltered by prying government hands. In fact, this is a concern of Tech UK also: “Over regulation of robotics and artificial intelligence at this stage of its development risks stalling or even stifling innovation. This could in turn risk the UK’s leadership in the development of these technologies.”
The government has a very important role to play in the development of artificial intelligence, though it has to appreciate it is a supporting role.
Creating regulation and policy to support the growth and prevent the abuse of AI should be the primary concern of public sector officials, though there are certain areas in the report where one could become concerns government hands are reaching too far.
The role of government is critical, that is not under question, though there has to be a careful balancing act to ensure actions are not at the detriment of progress. Only time will tell as to whether official bodies are able to keep clear of micro-managing industry, but it also can’t write a blank cheque to industry.