There are competing terms currently being decided on by the market at large:
AEO (Answer Engine Optimization) and GEO (Generative Engine Optimization)
Candidly I am working on a startup in this space myself, though we are taking a different angle than most incumbents.
While it's still early days for the space, I sense a lot of the original entrants who focus on, essentially, 'generate more content ideally with our paid tools' will run in to challenges as the general population has a pretty negative perception of 'AI Slop.' Doubly so when making purchasing decisions, hence the rise of influencers and popularity of reviews (though those are also in danger of sloppification).
There's an inevitable GIGO scenario if left unchecked IMO.
> There are competing terms currently being decided on by the market at large: AEO (Answer Engine Optimization) and GEO (Generative Engine Optimization)
It really annoys me the industry seems to be narrowing in on the two worse options rather than AIO.
For those not as well-versed in corporate PR....Salesforce are going to do just the bare minimum to keep the service going until the revenue dries up (or some > 0 $$ threshold where it just doesn't financially make sense to keep it running).
Pour one out for Heroku as they were truly a revelation back in the day and one of the most magical experiences ever on first run.
Somewhat related but I was disappointed to learn that Time Machine would no longer support Time Capsule post Tahoe, which I suppose is fair 8 years post discontinuation, but also unfortunate considering AFAICT the only real potential issue would be HDD degradation over the years? At the least, there have been plenty of system alerts noting this fact, but still annoying to have to buy something else. I know both AirPort and Time Capsule were an infinitesimally small part of their business...but they absolutely rocked when they were launched.
Amazon is generally good at 1) resolving an issue in your favor and 2) getting you to a human if needed but gosh does it feel like I've taken a different path to do so every single time I'ever needed support.
While totally unaware of the underlying economics, I do find it interesting how the major LLM providers found a way to get a non-trivial portion of consumers to actually pay for the consumer service. Of course, ads are still coming, but it was objectively impressive to go from 20+ years of "search is free" to "search is free, but capped, unless you pay us."
LLM use is not a simple search. I pay for it to either aid me or autonomously do work with document authoring, software development, and market research. It's not apples-to-apples when comparing.
if you looked at the underlying = economics - even a quick review - you'd see that paying customers is a relatively trivial portion. This is much closer to the dotcom race to maximum eyeballs; figure out the money part later.
Highly recommend adding some kind of canary like this in all LLM project instructions. I prefer my instructions to say 'always start output with an (uniquely decided by you) emoji' as it's easier to visually scan for one when reading a wall of LLM output, and use a different emoji per project because what's life without a little whim?
Does it actually? One sentence telling the agent to call me “Chris the human serviette” plus the times it calls me that is not going to add that much to the context. What kills the context IME is verbose logs with timestamps.
Sure, but its an instruction that applies and the model will consider fairly relevant in every single token. As an extremely example imagine instructing the llm to not use the letter E or to output only in French. Not as extreme but it probably does affect.
People are so concerned about preventing a bad result that they will sabotage it from a good result. Better to strive for the best it can give you and throw out the bad until it does.
Irrelevant nonsense can also poison the context. That's part of the magic formula behind AI psychosis victims... if you have some line noise mumbojumbo all the output afterward is more prone to be disordered.
I'd be wary of using any canary material that wouldn't be at home in the sort of work you're doing.
It is not a single emoji, it's an instruction to interleave conversation with some nonsense. It can only do harm. It won't help produce a better result and is questionable at preventing a bad one.
While it could stand to be more aggressive at times, especially at intersections, FSD works fairly well in NYC and can do all of less-than-legal-but-necessary things a normal driver can do (such as cross over a double yellow if there is a double parked car blocking the road) so I don't see why Waymo would have any trouble on that aspect at all.
reply