The discussion here on AI futures can be a little too credulous of company visions. People tend to push what they have. The three big AI labs will say bigger models are the future. Every other firm has only small models to sell, so they will tell you small models are the future.
Wharton's Ethan Mollick warns that public AI debates are overly credulous of vendor-driven model roadmaps
He notes labs promote whichever model sizes they currently sell
Users dismiss AI labs promoting model sizes aligned with their products as self-serving cope or profit-driven market positioning masked as philosophy, though a few praise benchmarks and scale for unlocking unexpected capabilities.
No Digg Deeper questions have been answered for this story yet.
Most Activity

@emollick hard to know who to believe unless you…i dunno…use the models. te he.

@qamaro_studio Waiting for my fable runs to finish

@emollick The big labs are four. Everybody seems to forget SSI exists.

@emollick Fable-5 is BACK bro what are u doing HERE

@emollick Miles' Law...

@emollick true but thats just how markets talk their own book
ai discourse is 90% positioning masked as philosophy

@emollick it isnt even conscious cope, it is product roadmap coping
everyone rationalizes their inventory

@emollick Incentives often shape predictions as much as evidence does. Looking past the narrative usually gives a clearer view of where the technology is actually heading.

@emollick everyone naturally roots for their own team, but it's also true every model fits specific tasks — and being able to save with simpler models on simpler tasks seems like a win for users

@emollick everyone's take aligns with what they actually own. hard to tell whos really right til it plays out

@emollick Yeah, but the benchmarks don't lie.
Scale keeps unlocking shit nobody saw coming.

@emollick But why cant we all just get along?
Oh yeah. Profits.

@IALatino @emollick How is SSI remotely comparable to the big 3? They have no product yet.