It means your whining about some gay guys acting inappropriately towards you
doesn’t somehow justify saying absolutely vile homophobic shit like “normal straight people get the willies when gay guys hit on them”,
does not somehow justify homophobia in general nor its conservative apologists and therefore
is incredibly tone deaf in this precise context.
That LLM is really pro-Macron, I wouldn’t take its opinion at face value.
The parts about defending democracy, addressing climate change, and preparing France for crises, in particular, are pretty ironic when you take into account the fact that both Amnesty International and the European Court of Human Rights have spoken up against his frankly appalling handling of the yellow vest crisis (the latter has a procedure against the French State dor acts of torture against demonstrators), that he’s been torpedoing our public health system even during the pandemic, and that his administration has been found guilty of “climatic inaction” by our own courts… The list goes on. For a while, “to Macron” in Ukrainian meant talking a lot while doing nothing. Maybe the LLM bought into the political communication.
That’s only true around landing and takeoff. For the most part their navigation relies on hybridized data from their inertial, air data and GPS, with several redundancies in place for bad readings and cumulative errors. Among all of this autonomous measurement apparatus, the GPS is the only part that doesn’t require numeric integration from speed or acceleration data to yield a position reading, and thus it is the only one that doesn’t drift over time. It’s actually fairly important, and it’s why using the gnss jammers you can find on amazon is super illegal
That’s absolutely true, generative AI is mostly a parlor trick with very few applications beyond placeholder art and faster replies to emails. But even for your kind of engineering problem, there’s still a big issue that’s often disregarded.
If we keep your example of an AI for a city grid, an important aspect of this type of engineering problem is guaranteeing that the system has as few catastrophic failures as possible (usually guaranteeing less than 1 for every 109 hours of uptime for systems where catastrophic means a certain quantity of dead bodies or high monetary costs, like a city grid, train signalization, flight control…). AI models may very well end up being discarded in those problems because even if you observe a better accuracy in simulations and experiments, mathematically proving this 109 figure is impossible because we don’t know how they work. Proving a threshold experimentally can happen, but a 109 number would require something like centuries of concurrent testing in every city in the world… I’ve just had a class with this example for trains. They were testing a system that reads signalization with a camera in order to move towards a more autonomous train. Deep learning performed better that classical image processing, but image processing allows you to prove that the train won’t misread less than x% of the time with way higher certainty than a black box, so they had to go with that if they ever wanted to pass safety certifications.
So I guess deep learning explainability might be a more significant challenge even that finding a dataset that isn’t racially biased…
I think he said fuck