Geek Stuff category archive

The Dot-Com Bubble v 2.0? El Reg reports that

Businesses are rethinking their AI plans in the face of changing cost structures and rising fees. The research also found nearly half of organizations have rephased AI deployments when costs have outweighed the expected value. Lower-cost, high-fidelity models are the fastest-growing influence on AI strategy, up 7 percentage points from Q1.

“These actions do not signal reduced confidence in AI. Rather, they suggest a growing willingness to evaluate where AI creates meaningful value and where it does not. Organizations appear increasingly focused on concentrating investment where expected returns are strongest,” the report said.

Much more at the link.

Capable of kidnapping your data? El Reg reports that (s)mooth AI criminal drives ‘first’ end-to-end agentic ransomware attack. (To put that another way, once it got the prompt, an AI bot ran a rensomeware scam from start to finish all on its ownsome.)

AI doesn’t stand for “ariticial intelligence.” It stands “amoral instrumentation.”

Impartial? Only as impartial as the data they scrape train on.

At the Psychology Today website, Phil Reed reminds that “AI mirrors human cognitive biases, not just factual knowledge.” Follow the link for the evidence.

A competent couples counselor? Better have that divorce lawyeer’s number handy.

Time of having “social” media decide what to hear about?

The EFF travels back into the past to recommend a way to escape tentacles of “social” media algorithms.

My old Philly DL Friend Noz tells a tale of the singularity.

A teaching tool? If you want more fools.

At the Psychology Today website, Soren Kaplan discusses a study involving 26,000 students indicating that “(u)sing AI can create short-term results but may stifle long-term learning.”

Impartial and unbiased? Just as impartial and unbiased as the internet it scrapes to fill its database: As Phil Reed warns us at the Psychology Today website, “AI in healthcare amplifies existing gender and cultural stereotypes, worsening inequality.”

Follow the link for his evidence.

(Slightly edited for clarity.)

From El Reg:

Waymo is recalling nearly 4,000 robotaxis after its vehicles repeatedly failed to recognize freeway construction zones, in some cases driving past closure signs or between cones marking closed lanes. . . .

Renenberm, robots are programmed by fallible human beings. Therefore, robots are fallib–no, never mind. Noone is listening and and everyone’s buying the hype.

At the Psychology Today website, Rosanna E. Guadagno argues that “social” media is by design, er, antisocial.

She makes three main points:

• Outrage drives engagement, so social media platforms amplify it by design.
• Reordering social media feeds alone shifted partisan warmth by more than 2 points in a 2025 field study.
• Disagreement is healthy, but contempt is what social media algorithms actually produce and reward.

Follow the link for some hints to avoid the artificially augmented algorithmic anger.

Seeing stuff that isn’t there? It’s, like, tripped out, dude.

A competent legal researcher? Maybe you should check with these lawyers.

A competent ER doctor? You may be headed for a quack-up.

Dressing up superficial slop in Sunday go-to-meeting clothes? Yes, argues John Nosta who warns us that (empasis added)

It’s important to recognize that slop isn’t simply bad writing. Bad writing is clumsy and unfinished and reflects the process of thinking. Slop operates differently. Its output is optimized for the read rather than the idea. Its purpose isn’t to explore a thought but to create the experience of having encountered one.