Note di Matteo

La storia si ripete:

Mr. Musk has reorganized xAI on the fly. He has led an aggressive recruiting drive for engineers. And he has pushed out a flurry of prominent researchers, even as others have left because they thought xAI had abandoned science in favor of attention-grabbing products, like a chatbot that sometimes produced offensive material and flirty A.I. chat companions, according to two people with knowledge of the company who spoke on the condition of anonymity.

(I capitoli precedenti sono la scissione di Anthropic per dissensi sulla scarsa attenzione alla ricerca sull'allineamento, e le grandi litigate tra OpenAI e Musk di nuovo sulla necessità di fare ricerca e in modo aperto: openai.com/elon-musk)

Scrive il WSJ che gli hard disk costituiscono l'80-90% dei sistemi di storage nei datacenter (la tecnologia SSD è troppo costosa), e con la necessità di raccogliere molti dati per il training di AI le aziende leader del settore (Seagate e Western Digital) sono in forte crescita.

E mi sorprende vedere che c'è ancora spazio per l'innovazione:

Both [companies] are moving toward a new hard-drive storage technology called heat-assisted magnetic recording, or HAMR. Seagate is rolling it out now, and Western Digital is set to do so in a couple of years.

HAMR opens the way to larger-capacity drives—30 terabytes and more. The companies have been working on the technology for many years, and there’s no easy path for competitors to challenge them.

I was at a leadership group, people were telling me they think that with AI they can replace all of our junior people in our company. That's one of the dumbest things I've ever heard. [...]

How's that going to work when ten years in the future you have no one that has learned anything. My view is you absolutely want to keep hiring kids out of college and teaching them the right ways to go build software and decompose problems and think about it, just as much as you ever have.

Matt Garman, CEO AWS (via)