Technology
This is the official technology community of Lemmy.ml for all news related to creation and use of technology, and to facilitate civil, meaningful discussion around it.
Ask in DM before posting product reviews or ads. All such posts otherwise are subject to removal.
Rules:
1: All Lemmy rules apply
2: Do not post low effort posts
3: NEVER post naziped*gore stuff
4: Always post article URLs or their archived version URLs as sources, NOT screenshots. Help the blind users.
5: personal rants of Big Tech CEOs like Elon Musk are unwelcome (does not include posts about their companies affecting wide range of people)
6: no advertisement posts unless verified as legitimate and non-exploitative/non-consumerist
7: crypto related posts, unless essential, are disallowed
view the rest of the comments
I was speaking generally. In other words, the LLM will convert 100% of what you tell it to but only part of the result will be correct. That's the problem.
And in this case they're not doing that:
So you might feed it your COBOL code and find it only coverts 40%.
I'm afraid you're completely missing my point.
The system gives you a recommendation: that has a 50% chance of being correct.
Let's say the system recommends converting 40% of the code base.
The system converts 40% of the code base. 50% of the converted result is correct.
50% is a random number picked out of thin air. The point is that what you end up with has a good chance of being incorrect and all the problems I mentioned originally apply.
One would hope that IBM's selling a product that has a higher success rate than a coinflip, but the real question is long-term project cost. Given the example of a $700 million dollar project, how much does AI need to convert successfully before it pays for itself? If we end up with 20% of the original project successfully done by AI, that's massive savings.
The software's only going to get better, and in spite of how lucrative a COBOL career is, we don't exactly see a sharp increase in COBOL devs coming out of schools. We either start coming up with viable ways to move on from this language or we admit it's too essential to ever be forgotten and mandate every CompSci student learn it before graduating.
Again, my point really doesn't have anything to do with specific percentages. The point is that if some percentage of it is broken you aren't going to know exactly which parts. Sure, some problems might be obvious but some might be very rare edge cases.
If 99% of my program works, the remaining 1% might be enough to not only make the program useless but actively harmful.
Evaluating which parts are broken is also not easy. I mean, if there was already someone who understood the whole system intimately and was an expert then you wouldn't really need to rely on AI to port it.
Anyway, I'm not saying it's impossible, or necessary not going to be worth it. Just that it is not an easy thing to make successful as an overall benefit. Also, issues like "some 1 in 100,000 edge case didn't get handle successfully" are very hard to quantify since you don't really know about those problems in advance, they aren't apparent, the effects can be subtle and occur much later.
Kind of like burning petroleum. Free energy, sounds great! Just as long as you don't count all side effects of extracting, refining and burning it.
A random outcome isn't flipping a coin, it's rolling dice