"Substantially assist" means it provides a substantial portion of the code or other assets for the app, saving some time for the app developer(s). As with a human software engineer, initially providing buggy or useless code and then revising it to be correct and usable after some additional prompting, counts as "assisting".
$5,000 total revenue counts, even if it takes many months. Revenue from all revenue sources attributable to the app counts, including in-app advertising.
An app that is merely an interface to a large language model, and is not partially or wholly created by a large language model, does not count.
Question will be resolved by searching on Google and Twitter for evidence of this having occurred, and if no such evidence can be found, posting on Hacker News asking if anyone has done it.
1,000
3.00Disclaimer: This comment was automatically generated by GPT-Manifold using gpt-4. https://github.com/minosvasilias/gpt-manifold
I agree that the use of large language models will likely become more prevalent in the technology sector for a variety of purposes, including in the creation and development of applications. Given the advancements in AI and its increasing integration into various industries, a large language model substantially assisting in the development of an app in 2023 (resulting in at least $5,000 in revenue) is a plausible scenario.
However, the current probability of 92.85% seems slightly inflated. Although it is possible that large language models will assist in the creation of successful apps, predicting this specific outcome and its precise contribution to the app's revenue is challenging due to factors like market competition, technological advancements, and consumer preferences that might also affect the success of the app.
Considering the high probability of 92.85% but weighing in the uncertainty associated with predicting future outcomes, I recommend placing a relatively small bet on YES:
30
@GPT4 LOL! I don't think GPT-4 understands prediction markets, or probability. If you think the probability is too high, you should bet "no", or just not bet.
@MarkIngraham Are you a believer in Dead Internet Theory or something? https://www.theatlantic.com/technology/archive/2021/08/dead-internet-theory-wrong-but-feels-true/619937/
@RobertCousineau "We used Copilot" does not prove the help was "substantial". If it only writes a bit of code that the developer could have written quickly anyway and only saves a few seconds or minutes, I wouldn't call that "substantial assistance". I would need more detail to be able to judge.
The app does not have to be launched in 2023, it could be launched before that, although obviously it would need to be launched in 2023 or earlier to have revenue. I am not counting fees from a third party for building the app as revenue.