Thread regarding Meta / Facebook layoffs

Panic in the disco

meta aint in no zombie era…its in panic mode. The times says meta is turning into Aol or yahoo, some old big internet comapny that used to be on top and now nobody cares. i beleive they kinda right, but they missing the real thing. meta dont know what it is no more. first zuck changed the whole name for metaverse and burned like $80b on reality labs. they told everybody vr meetings and cartoon avatars was the future, but nobody wanted it. horizon worlds had tiny numbers, avatars had no legs, and the whole thing looked d-mb. then ai got hot and meta definately jumped on that too, like they just chase whatever is shiny.

now they getting sued by publishers and an author for allegedly using pirated books to train ai. lawsuit says they knew the data was pirated and used it anyway. it also says zuck ki-led licensing talks cause if they pay to recieve even one book then it hurts their fair use story. so this aint seperate problems. its all one mess, wich is a buisness with no plan, just shortcuts and money burning. then u got reports of meta watching employee keystrokes and mouse movement to train ai agents that might replace them later. maybe that occured cause leadership is scared. no suprise regular workers get layoffs, stack ranking, rto pressure, bad morale, and fear while the top keeps funding zucks next fantasy.


by
| 1 view | | 4 replies (last 21 days ago) | Reply
Post ID: @OP+1ks716eqr

4 replies (most recent on top)

@cq
explain me,
random key stroke on keyboard, by 8000 people , and train AI and make meaning full MVP

by
| | Reply
Post ID: @d6+1ks716eqr

OP, so facebook is the MySpace of the 2020s? Meta's products are a blight on society. The sooner they go away the better.

by
| | Reply
Post ID: @cr+1ks716eqr

@ck what is your expertise in training AI models? How many have you help develop?

by
| | Reply
Post ID: @cq+1ks716eqr

The idea of tracking employee keystrokes and mouse movements to train AI agents fundamentally misunderstands how humans solve problems.

There is no linear pattern to human thought. When a problem appears in front of us, our workflow is messy: we constantly iterate, undo, change our rationale mid-way through a task, and continuously pivot until we find a working solution. Even then, we often go back to improve it when a better idea strikes.

Raw activity data only captures the mechanical execution, not the underlying cognitive process. How can tracking a cursor or counting keystrokes possibly train an AI to understand the logic, intuition, and shifting strategy behind human decision-making?

d-mb to capture keystroke to learn,
maybe to identify if you are actually working makes sense.

by
| | Reply
Post ID: @ck+1ks716eqr

Post a reply

: