When John Barber developed the first gas turbine in 1791, he could have never imagined the transformation his invention would bring.
By 1886, after almost a century of technological innovation, Karl Benz started commercial production of the first automobile with an internal combustion engine.
A decade later, one of his products was run over by Brigitte Driscoll near Crystal Palace, making the Croydon resident the first British person to be killed by a car.
Drivers of the day did not require a license, and licenses were introduced until 1903, after which they were not required to pass a driving test for 32 years.
By that time there was a Highway Code (first promulgated in 1931) containing instructions for horse-drawn carriages, and almost 30 years before motorists had similar rules, it was illegal to drive while intoxicated. bottom.
Seat belts, which cut fatalities in half, were not mandatory in the UK until 1983.
The point is that while the internal combustion engine was a world-changing technology that took decades to develop, effective regulation took even longer.
As manufacturers moved ahead and the pavement economy transformed, lawmakers choked in the dust like red-flag men paid to walk in front of new machines.
AI revolution
today artificial intelligence A similar industrial revolution is promised, but the pace of development can be measured in months and years rather than decades, and its inventors have a clear eye for risk.
Last month saw apocalyptic predictions, with major developers predicting that “generative” AI, with the ability to generate text and images from prompts and learn in the process, could become a “social force” similar to a pandemic or nuclear war. warned that it poses a risk of scale.
Regulation and oversight have therefore become important, with Rishi Sunak saying the sector is one he would like to own, declaring that the UK could lead the international debate.
He plans to host a “world summit” in the fall and propose a possible follow-up by a UK-based body modeled after the International Atomic Energy Association.
AI that collides like a “freight train”
The move is part of a broader ambition to not only provide positive coverage of the prime minister’s visit to Washington, but to position the UK as a hub for AI and make digital innovation a priority for delivering growth.
He also sees regulation as an opportunity, but it’s not so clear how that will actually play out.
We know that as recently as March the government had intentions of following the car model.
In the white paper, the company said it would focus on “the use of AI rather than the technology itself” in order to “make the responsible application of AI flourish.”
Regulation of AI
Instead of passing laws restricting the technology, existing regulators will monitor its application in the field and work with developers to establish acceptable boundaries.
So rather than create a central authority for AI, health regulators will oversee the use of AI in diagnostics, Ofcom will continue to crack down on machine-generated online misinformation, and roads and railroads will inspect transportation infrastructure. will be responsible for monitoring whether it is safe for AI to analyze
This model has already been successfully applied in industries where generative AI is used.
Energy company Octopus uses AI tools to answer more than 40% of its customer interactions, but to comply with data protection laws, it removes all personal data from emails before AI reads them. is deleting.
In recent weeks, Mr. Sunak has gone further, talking about the need for “guardrails” to control AI, but there are concerns that regulators are already lagging behind.
The Trade Union Congress (TUC) believes tougher re-employment laws are already needed. AI is already being used by employers to screen job applications, and in some cases unions believe they make hiring and firing decisions.
They argue that everyone should have the legal right to appeal decisions to humans, rather than relying on machine judgment, especially as AI learns from and leverages past experience, as prejudices and prejudices can take root. I hope to have
Introducing the world’s first ultra-realistic robot
The UK’s approach contrasts with that of the EU. The European Commission is proposing what is believed to be the world’s first legal framework for AI, based on four levels of risk to people’s lives, safety and rights.
The implementation of social scoring by governments and the use of AI in toys that may encourage unsafe behavior are considered “unacceptable risks” and outlawed.
Areas of minimal risk include video games and spam filters, while limited risks include the use of chatbots, as long as it is clear that you are talking to a machine.
High-risk areas include any application such as education, critical infrastructure, employment such as resume sorting, immigration decisions, public sector decision-making, and the justice system.
For AI tools to be legal in these fields, they must meet a number of conditions, including “appropriate human oversight” and the ability to track and record activity.
read more:
Who are the “Godfathers of AI”?
Smartphone camera lens technology used to diagnose thousands of patients
When Computers Say No, We Need to Know Why
The challenge for developers and users today is that it is not always clear how the AI reached its conclusions.
When ChatGPT or any other language tool produces plausible human-generated text, you cannot know where that information or inspiration came from.
It may not even matter if you ask me to write you a limerick or a letter.
It is very important if it determines whether you are eligible for benefits. If your computer says no, you need to know why.
Making the UK the home of technology regulation, a kind of digital Switzerland, is an attractive post-Brexit ambition, but whether it is possible is debatable.
As the ongoing fuss over maintaining EU law shows, we may want to create our own regulations, but commercial logic may force us to follow the larger market. .
However, doing nothing is not an option.
AI is advancing so fast that the UK cannot afford to be left by the roadside as it heads towards the horizon.
Uniquely, it may be the first technology that knows more about its destination than we do, for better or worse.
