The prime minister of Albania, Edi Rama, has appointed the world’s first AI-generated minister to tackle state corruption and aid innovation in his cabinet. This was inevitable, especially in small corrupt nations. I know my own South Africa could really benefit from this anti-corruption AI tool, though here someone will figure out how to swindle even the AI.
Nations fall due to the corruption of middle management, if not the fat cats at the top. The constitution may be solid, the bright ideas timeously adopted, but if the people in positions of power are greedy and selfish, you have no chance of success. I have seen it happen here like clockwork, the decay by eating away the state institutions from inside like parasites.
The AI minister, developed in partnership with Microsoft sounds like a good solution, but can we trust Microsoft? I wonder about that.
Rama said that the AI-generated bot would ensure that public tenders were “completely free of corruption”. If it works there, we'll see it roll out everywhere. And soon the president of the world will be an AI robot, because humans are fallible. All of us. At least in the eyes of the opponent.
So who will program the robot to run the world? Who will watch the watchers? This is so risky that some of the top AI developers in the world have their P - doom at 50%. They believe that the chance of AI taking over and destroying things is 50%. That's atrocious odds.
Historically, government procurement in Albania has been vulnerable to bribes, threats and vested interest. As it is everywhere, though it's just better hidden in other nations. Of course theoretically things like smart contacts in blockchain and AI robots are meant to protect and improve our systems, but they too are fallible.
A multi-tool suite known as Humphrey was launched by the UK government earlier this year. It was named after a fictional bureaucrat from the 1980’s British sitcom Yes Minister. So here we go.
Meanwhile in France, then prime minister Gabriel Attal launched a French-made AI system known as Albert in April to streamline administrative procedures. This is where it starts. A little secretary on the side to speed life up a bit. Next thing AI is running the war on the battle field for us. And that battle field is also in cyberspace, isn't it?
I'm not against AI LLMs and robots. I believe they are inevitable. I use Chatbots whenever I like. They are free, some are open source.
Yet it looks so dangerous when used for big issues like political decisions. We'll probably make a few mistakes along the way as a civilization, as we move fast and break things at our current pace. The problem is the tool is sharper than ever and the consequences of a mistake in the wrong hands could be global. So the stakes are higher than ever.
Image: https://pixabay.com/photos/transformers-optimus-prime-car-1333083/