Not enough time to really change.
-0.028988
Relative Brier Score
2
Forecasts
0
Upvotes
Forecasting Activity
Forecasting Calendar
No forecasts in the past 3 months
| Past Week | Past Month | Past Year | This Season | All Time | |
|---|---|---|---|---|---|
| Forecasts | 0 | 0 | 2 | 2 | 2 |
| Comments | 0 | 0 | 1 | 1 | 1 |
| Questions Forecasted | 0 | 0 | 2 | 2 | 2 |
| Upvotes on Comments By This User | 0 | 0 | 0 | 0 | 0 |
| Definitions | |||||
Most Active Topics:
International Diplomacy & Conflict,
Science & Technology,
Iran: Threats & Influence
Most Active Topics:
Decoding Disinformation,
Iran-VNSAs
New Prediction
New Prediction
Probability
Answer
70%
Less than $90
30%
More than or equal to $90 but less than $100
0%
More than or equal to $100 but less than $110
0%
More than or equal to $110 but less than $120
0%
More than or equal to $120
Files
New Badge
Why do you think you're right?
I don't believe twitter is likely to do anything about AI generated posts any time soon. No articles on research for it, and their AI generated photo world is purely based on user feedback.
Meta (Facebook, whatsapp and messenger) are all the same parent company. They've already got something in place for AI generated photos. The technology is similar, though harder to do with posts.
They were quick to add disinformation abilities, and I believe that in an election year in the US they will want to be quick about getting ahead of the AI posts. The issue is how far / long it takes to train the model to see AI generated posts. So erring on the side of unrealistic to see it by June. Though I would expect to see more media / mention of them working on it before long.
Why might you be wrong?
Twitter is a dark horse. Just because they haven't talked about it doesn't mean they aren't working on it.
And for Meta, using exsiting tools to train the model faster could speed up the process. With the AI image tags already in place, they have the ability to shift more focus on text now and that could speed up the process.
You may have missed their president of global affairs, Nick Clegg, stating recently that "that ship has sailed" regarding the possibility of labeling AI-generated text, and that there was currently no viable mechanism to label written text generated by AI tools like ChatGPT: https://www.infer-pub.com/comments/118139