Why does the AI need SQL queries? Who needs that? It will just write its own ACID-compliant database with its own language, and while it's at it, reinvent the operating system as well. It's turtles all the way down.
It’s actually not a ridiculous concept, but I think in some ways code will go away, and the agent itself will do what the code used to do. Software of the future will be far more dynamic and on the fly. We won’t have these rigid structures
I use a similar strategy for API design. Every API call is wrapped in a large database transaction, and I either roll back or commit the transaction based on dry-run or wet-run flags. This works well as long as you don’t need to touch the file system. I even wrap emails this way—emails are first written to a database queue, and an external process picks them up every few seconds.
The code is not littered with dry-run flag checks; the internal code doesn’t even know that a dry run is possible. Everything is rolled back at the end if needed.
All database referential integrity checks run correctly.
Some drawbacks: any audit logging should run in a separate transaction if you want to log dry runs.
I've just got a new Samsung and it's wobbling too. I hate this. Why can't they at least put the cameras in the middle? Or maybe horizontally centred? Or they could just put another bumper on the other side to make it symmetrical. I'm looking for a cover to balance this out, but no luck so far.
When I started high school in the early '90s, there was a compulsory summer reading list of 10–12 books, each ranging from 300–800 pages. Then we had to write essays about them. This was just our summer homework before the new school year started. I didn't enjoy it at all; at the same time, I read lots of easy fiction, sometimes several hundred pages a day.
My six year old (who is still in kindergarten) reads about 70–100 pages per week of books aimed at eight to nine year olds.
Just two days ago, I asked ChatGPT to provide an explanation of the place-value system that my six-year-old could understand. The only problem was that it mixed up digit value and place value, which caused it to become confused. I spotted the mistake, and ChatGPT apologised, as it usually does. But if my six-year-old had asked it first, she wouldn't have noticed.
I'm not sure how much misinformation my child would learn as truth from this device.
For me, it seems the issue with the recipe was due to pre-recorded segments. The guy quickly realized that the "AI" was ahead of the script and cut it short to avoid making a bigger mess during the demo.
As for the Zuck segment, I believe it was a race condition.
The last time I asked for a code review from AI was last week. It added (hallucinated) some extra lines to the code and then marked them as buggy. Yes, it beats humans at coding — great!
