Weekly Blog - 1 May 2023 - Artificial Intelligence
Artificial Intelligence
Warnings about Artificial Intelligence (AI) are in the news everywhere this week. On Monday 1 May the New York Times carried an interview with computer scientist Geoffrey Hinton, the ‘Godfather of AI’, who recently quit Google warning of the dangers of AI. The speed of development of AI has increased dramatically in recent months as companies around the world race to get ahead of the chatbot ChatGPT, which was launched in November 2022. ChatGPT already has over 100 million active users, and has been able to deliver astonishing, detailed, largely accurate and complex answers to virtually all questions. In March more than a 100 industry experts including Elon Musk signed an open letter calling for a pause in AI development because of the risks. Meanwhile the White House has begun a public consultation on how AI should be regulated.[1]
The risks and the opportunities
The risks from AI are numerous. A huge range of jobs in many different industries could be lost as AI is able to produce better and faster answers to multiple complex questions, tasks and even creative output. The access to online data that AI requires also raises important questions about data security. The underlying algorithms used by AI means it can learn and get smarter and smarter, very rapidly over time. They also lead to the increasing likelihood that relatively simple underlying instructions give rise to more complex, perverse and unwanted behaviour from AI as it learns and interprets those instructions in unpredictable ways. And ultimately, beyond all of that lies the risk of genuine consciousness emerging.
On the other hand, there are also many potential opportunities presented by AI. Many jobs that are boring and repetitive may be replaced, releasing people for more interesting, rewarding and higher paid work. Aspects of many other roles can be done better by AI, which is in many cases already more knowledgeable than any human, more accurate, able to make better decisions, and produce better results. Beyond this, new economic opportunities and business models will open up, and overall knowledge will rise to a higher level, along with other benefits for humanity. So what might be a balanced approach to AI, that neither ignores the risks or supresses the opportunities?
A balanced approach
The global economy is always changing and evolving as new technologies like AI emerge. This is a key finding in a major research report from Arise, the Arise Manifesto, which looks at Biblical teaching and all the lessons from history to develop a Christian vision for a better world in the five major areas of evangelism, discipleship, social justice, development and the environment. Throughout human history, the dynamics of this process of technological and social evolution have been essentially simple: different people in different places have invented and tried new things to improve their lives. Sometimes these innovations have had a single point of origin, at other times, similar ideas have been tried in varying locations independently. Ideas that have worked have been adopted and copied by others. This same slow and gradual process has taken us from the first stone tools to the latest online technology, and from Magna Carta to today’s complex international legal codes. As the historian Ian Morris puts it in his book Why the West Rules – For Now, “For millennia social development has generally been increasing, thanks to our tinkering, and has generally done so at an accelerating rate. Good ideas beget more good ideas, and having once had good ideas we tend not to forget them.” Or put less generously, “Change is caused by lazy, greedy, frightened people looking for easier, more profitable, and safer ways to do things. And they rarely know what they’re doing.” [2] (Arise Manifesto, pg 166)
Through this process, whenever new technologies like AI emerge, some jobs in old industries based around old technologies will come to an end. However, at the same time those new technologies will be opening up new jobs and industries. The key role for governments is to ease the pain of this transition, so that those whose jobs are lost, are quickly retrained to find new employment in the new industries that are opening up. Governments should also provide a social safety net of basic services to support communities through this process of transition (Arise Manifesto, pg 186).
Having said that, another key learning from the Arise Manifesto is that as new science and technology evolves this cannot just be left to happen in a lawless Wild West fashion, and society left to pick up the damage. Rather there should be some public debate, investment into research and development in technology that advances society in socially positive directions, and legislation which prohibits scientific and technological research in directions that are dangerous or bad for society (Arise Manifesto, pg 260 – 261).
In this context, a temporary pause on the further development of AI to enable this debate and scrutiny to happen (and legislation put in place to mitigate and avoid the risks) does seem like a sensible idea. An international agreement for a temporary pause should be relatively easy to agree. It is in every nation’s interest. Quick but detailed discussions could take place, and guidelines, agreements and legislation drawn up to enable things to move forward more safely. Even if a universal agreement can’t be reached, an agreement between at least the democratic members of the international community, or the OECD members, should be relatively straightforward. The world has taken this kind of rapid internationally coordinated action in similar moments of risk in our recent past. Covid 19 in 2020, and the G20 coming together to quickly prevent the unravelling of the global financial system during the financial crisis in 2008 – 9 are recent examples. There are precedents. Global problems that impact on every country, require nations to come together and reach global solutions to deal with them (Arise Manifesto, pg 272 – 274).
It seems that Artificial Intelligence in one form or another is here to stay. It can’t be uninvented. The genie is out of the bottle. But, it is very much in the power of society and governments to exercise caution and oversight, and put good rules and safeguards in place to ensure we get the upsides from AI, and avoid the risks.
Find out more
Find out more about how God is at work in the world, and the role we all have to play in that work, in the Arise Manifesto. This report is Arise’s big picture, researched, Biblical, holistic and practical vision for a better world. It looks at what the Bible says, and what we can learn from the best data and the world’s leading experts on the five major areas of evangelism, discipleship, social justice, development and the environment. It then draws these lessons together into a practical road map for the changes we need to see in our world, which the Arise movement campaigns to achieve.
Found this blog online, or sent it by a friend? Sign up to receive weekly blogs from Arise directly.
[1] Geoffrey Hinton: Who is the ‘Godfather of AI’?, Sky News, (2 May 2023), https://news.sky.com/story/geoffrey-hinton-who-is-the-godfather-of-ai-12871205
[2] Morris, I., Why the West Rules – For Now, (London: Profile Books, 2011), pp. 27 – 28