Google execs say employees have to 'be more AI-savvy'

12 mfiguiere 11 7/29/2025, 10:33:40 PM cnbc.com ↗

Comments (11)

bvan · 10h ago
Some of us have seen a couple of past AI waves and plenty other tech waves. Show me the money.. in the meantime we’ll get on with our jobs and wait for the other shoe to maybe drop.
flembat · 4h ago
The execs need to try actually using AI and verifying the correctness or lack of it in the results, THEY need to be more savvy about what AI is able to do. And if the results are garbage, and it wastes time rather than saves time they must be using it wrong, right?
pclowes · 8h ago
Bluntly I have my doubts on how "AI-savvy" the Google execs are based on how flat footed they seem to be with respect to releasing successful and compelling AI products.

Eg. I don't want Gemini to summarize a google doc, thats what the title and first paragraph do...

s1mplicissimus · 10h ago
I suggest replacing execs with LLMs just to make sure we get the maximum value out of the technology
blinkbat · 10h ago
they'd do a great job, and faster to boot, right?
s1mplicissimus · 10h ago
yes, and waaay cheaper
sovietmudkipz · 10h ago
Enterprise software engineers have to be good with AI to know when it will save time deploying it.

Understanding where it’s effective and how to use AI tooling is a sought after skill.

I would advise programmers to understand an AI tooling stack that they know very well.

blinkbat · 10h ago
give me an example of this "tooling stack" that isn't just prompting the LLM, or using a Cursor-like or Claude CLI-like?
blinkbat · 10h ago
if I was working for google I'd resign. supremely irresponsible for them to push AI this hard, it is simply not capable of replacing engineers nor speeding them up considerably if they have to review all of its output. seems like all the devs will just be pushing AI slop into the repos and letting the poor fucker in charge of reviewing it take all the inevitable heat

unless their tooling is somehow lightyears ahead of market and doesn't constantly bullshit or write broken, insecure code...

dekhn · 10h ago
i got a slow start on using coding agents but dove into it recently and it's allowed me to make more progress on some of my longest-term projects than I have in years! I also used to work for Google and this tool would have made my life tremendously better. I originally expected the coding agent to mostly handle boilerplate, but it produces initial prototypes that are often close to complete.
hustwindmaple · 8h ago
Any decent company would have comprehensive tests and peer review in place to mitigate AI bs