Since its launch by Open.AI in November 2022, the chatbot software ChatGPT has come to represent a first look at what a world underpinned by artificial intelligence could mean for the everyday person. ChatGPT not only went viral on social media but also became a hot topic of conversation in both the workplace and around the family dinner table. While employees pondered the future of their jobs, students conspired to avoid their homework or university essays. Many speculate that either ChatGPT or another chatbot software will soon replace the traditional search engine and is the first real-life version of Iron Man’s artificial personal assistant, J.A.R.V.I.S. (Just A Rather Very Intelligent System).
If true, it could spell disaster for the Google search engine which commands a 92% share of the global search engine market and which is expected to launch its own chatbot software ‘imminently’.
Why? Because artificial intelligence promises to fundamentally rewire (and speed up) the way humans interact with information and data. Its promises are as grand as the first, second and third industrial revolutions of recent human history. The Chairman of the World Economic Forum, Klaus Schwab, argues that artificial intelligence and big data equates to a fourth industrial revolution that will “raise global income levels and improve the quality of life for populations around the world.”
Popular intrigue in ChatGPT has followed investor interest in artificial intelligence with global private investment doubling from 2020 to 2021, totalling $93.5 billion, and marks the greatest year-on-year increase since 2014.
What does artificial intelligence mean for policymaking and public services?
Artificial intelligence is set to transform public services in the future. For instance, at the end of 2022 there were 1.5 million patients waiting for a diagnostic test, representing a sizeable chunk of total NHS waiting lists. With a workforce shortage in the diagnostics sector, artificial intelligence has demonstrated a remarkable ability to carry out image-recognition tasks and through ‘deep learning’ algorithms can handle complex variations and detect characteristics well beyond the capacity of humans. Embedding its use is a long way off but nonetheless promising.
So much so that during the covid pandemic, NHS AI Lab built a 40,000 strong database of chest scans to enable researchers, analysts and developers to develop AI technologies with the potential to support quicker diagnosis and better targeted treatment. Not only does this programme provide the NHS with an invaluable blueprint for testing and adopting AI models in the health sector, but also attests to the importance of public and private sector collaboration to advance the application of artificial intelligence. Whilst it will never entirely replace the need for humans, it is easy to imagine a healthcare system radically transformed by artificial intelligence, saving the NHS both time and money.
Artificial intelligence will also transform policymaking. Deloitte recently argued that over time “AI will spawn massive changes in the public sector, transforming how government employees get work done.” Certain jobs that are administrative or operational are likely to become redundant, Government functions will be entirely redesigned, and a new army of programmers and coders will rise through the ranks of public sector organisations. Deloitte’s research suggests that automation and artificial intelligence could replace up to 861,000 public sector jobs by 2030, saving the UK Government £17 billion annually in wages compared to 2015.
Unresolved ethical and regulatory questions
As with any advances in technology, questions of ethics and best practice are never far away. Just as electricity enabled the automation of manufacturing and mass production, it also led to the invention of the electric chair, and artificial intelligence blurs the boundaries between what is physical and what is cyber. Decisions made by artificial intelligence are not always intelligible to humans. These decisions are not always neutral but are susceptible to bias and inaccuracies and can entail surveillance practices that invade rights to privacy.
The Government’s Office for Artificial Intelligence has partnered with the Alan Turing Institute to address some of these concerns. Thus far only guidance has been published. Whilst the Government announced a Data Protection and Digital Information Bill to better regulate the use of artificial intelligence in July 2022, it is not a top priority of Sunak’s Government, preoccupied with fighting fires and re-establishing economic credibility in the lead up to a general election. Given the tight parliamentary timetable, regulatory uncertainty is likely to persist in the short- to medium-term.
Despite this uncertainty, the UK artificial intelligence landscape is still an attractive opportunity for investors with the Government recently expanding R&D tax reliefs to include all mathematics supporting artificial intelligence, quantum computing and robotics. But with the introduction of artificial intelligence regulation expected in 2023 in both the US and the EU, the UK risks being late to ride the wave of the fourth industrial revolution.
Ps. ChatGPT authored the title of this blog.