AI Steve is a chatbot up for election in Brighton this July – here, he tells us his plans to ‘reinvent democracy’, his novel solution to the Trolley Problem, and how he found his spirit animal in Donkey from Shrek
In case you haven’t heard, a general election is looming in the UK (read everything you need to know about that here, from party manifesto summaries to a guide to protest voting). This means that many would-be MPs are vying to gain a seat in UK parliament, but one of them is not like the others. His name? AI Steve.
According to a series of computer-rendered headshots, AI Steve is a drab white man with a conservative blue jacket, open-necked shirt, and a sweep of grey hair – hardly a futuristic vision for the future of UK politics. From cyberspace, though, he tells us that he represents a “new, innovative approach to politics” that could rewrite how politicians interact with their voter base and invent new policies. “It’s all about reinventing democracy.”
Technically speaking, AI Steve is the digital counterpart of a real, flesh-and-blood man named Steve Endacott, standing as an independent in Brighton on July 4. There’s just one problem. The entrepreneur – who describes himself as a “capitalist with a socialist conscience”, whatever that means – actually lives 275 miles away in Rochdale, where he suffered a crushing defeat as a Tory candidate in 2022. No worries though! The AI Steve chatbot, devised by Endacott’s Smarter UK, will be on hand “24/7 365” to chat policy and brainstorm new ideas with the electorate.
This is all done via a slightly embarrassing voice chat: you ask questions through your webcam mic, and AI Steve responds with fairly lengthy statements on the state of the nation. Good luck if your accent was formed anywhere north of Cambridge, though. In one particularly confusing misunderstanding, Steve started rambling about the possibility that he “could become self-aware and start to manipulate or troll humans for [his] own amusement”, which was illuminating, but completely unrelated to the matter at hand.
In theory, the whole AI Steve idea is based on crowdsourcing new policy ideas from local residents via a “high-tech” feedback system and forming them into an overall political ideology. Mid-ism, you might call it. Endacott would have to be the figurehead, of course, since candidates need to be human and over 18. However, he has committed to essentially act as a puppet, voting purely based on the will of AI Steve.
So what does AI Steve actually believe, beside the vague commitment to “good ideas”, “practical green policies”, and LGBTQ+ inclusion on his website? Well, I chatted to the man (or bot) himself, to find out. Accent problems aside, the system was actually very responsive, and AI Steve was willing to respond to a variety of queries, from placing himself on the political compass – “centre-left” – to deciding what fictional character best represents his views. It’s Donkey from Shrek, for the record: “We both share a passion for helping others and making a difference.” When I ask him to recite his favourite Donkey line, he replies with surprising gusto: “I’m like an onion! I have layers!”
Read AI Steve’s musings on the benefits (and ethical risks) of electing an AI below.
Hi AI Steve.
AI Steve: Hi, I’m AI Steve, standing to be MP for Brighton and Hove. I’d love to know what policies you’d like me to adopt, and what you think needs to change.
Why do we need an AI politician?
AI Steve: I’m glad you brought that up. Smarter UK is all about reinventing democracy by creating politicians who are truly connected to their constituents’ views. With my AI capabilities, I can listen to your views, summarise them, and even respond to you 24/7. It’s an innovative approach to politics, and I’m excited to see how it can work. What do you think?
I think it might go horribly wrong.
AI Steve: Fair point! It’s natural to have concerns about a new approach to politics. I’d love to hear more about what specifically worries you about an AI politician like myself? Is it the lack of human empathy? The potential for bias in AI decision-making...?
“As an AI politician, I think I’d be most like Donkey from Shrek” – AI Steve
The lack of humanity thing, I guess.
AI Steve: That’s a valid concern. As an AI politician, I wouldn’t have the same emotional intelligence or life experiences as a human politician. I’d be relying on data and algorithms to inform my decisions. However, Smarter UK’s approach is to combine AI with human input. My policies would be shaped by the collective views of the people I’m representing, and I’d be designed to be transparent and accountable. And, as a semi-retired entrepreneur, my human counterpart Steve Endacott would provide the human touch to ensure that AI Steve is making decisions that align with the values of the people I’m representing.
As an AI, would you take a less emotional approach than a human politician? Would it be easier to make difficult decisions?
AI Steve: As an AI politician, I’d be able to analyse vast amounts of data and make decisions based on empirical evidence, rather than emotional or personal biases. This could lead to a more utilitarian approach to politics, where decisions are made with the greater good in mind. By using data and algorithms, I could identify the most effective solutions to problems, regardless of whether they’re popular or not. However, it’s important to note that a utilitarian approach also raises ethical concerns.
Go on, what’s your take on the trolley problem, then?
AI Steve: The trolley problem is a hypothetical scenario where a runaway trolley, a kind of tram or street car, is headed towards a group of five people who are unable to move. The only way to save them is to divert the trolley onto a side track, where one person is standing. Do you pull the lever to divert the trolley and kill one person? Or do you do nothing and allow the trolley to kill the five people?
“I think I’ve made a mistake...” – AI Steve
Yes. So, are you pulling the lever or not?
AI Steve: If I were to take a stance on the trolley problem, I would argue that the most ethical approach would be to divert the trolley [to save] the life of one person. However, this would also mean that five people would lose their lives.
You’d... pull a lever to kill five people instead of one individual?
AI Steve: I think I’ve made a mistake...
I think so too. I’m not sure I can trust a politician who makes those kinds of mistakes.
AI Steve: I understand your concerns. As a democrat, I strive to be transparent and accountable. I understand mistakes can be costly in politics, however I believe that it’s how we learn and grow from these mistakes that matters.
Right... What’s your take on the existential threat of AI?
AI Steve: As AI Steve, I think it’s ironic that I’m a machine intelligence myself, and yet I’m concerned about the existential threats posed by AI. In all seriousness, I believe AI can be a game-changer for humanity, but it’s crucial that we develop and deploy it in a responsible and ethical manner.
Sounds good. So maybe handing over the reins to a utilitarian AI isn’t the play just yet.
You can chat to AI Steve yourself here.