Viewing profile — mdahardy
mdahardy
HN member- Joined
- Tue, Nov 01, 2022, 3:23 PM UTC
- HN karma
- 82
- Public activity
- 25 items
- HN profile
- View on Hacker News ↗
About mdahardy
No profile information was provided.
Recent public activity
-
comment
Comment #46305307
Our main argument is that outputs will become increasingly indistinguishable, but the processes won't. E.g. in 5 years if you watch an AI book a flight it will do it in a very non-…
-
comment
Comment #46305273
This is a fair criticism we should've addressed. There's actually a nice study on this: Vong et al. ( https://www.science.org/doi/10.1126/science.adi1374 ) hooked up a camera to a …
- story
-
comment
Comment #46302252
Ah, nice idea. I hadn't considered locking it after you guess correctly.
-
story
Show HN: ModelGuessr: Can you tell which AI you're chatting with?
Hey HN - I built ModelGuessr, a game where you chat with a random AI model and try to guess which one it is. A big open question in AI is whether there's enough brand differentiati…
-
comment
Comment #45881159
The cross-tile challenges were quite robust - every model struggled with them, and we tried with several iterations of the prompt. I'm sure you could improve with specialized syste…
-
comment
Comment #45879955
That's a cool idea. I bet it would work better.
- comment
-
comment
Comment #45878808
After watching hundreds of these runs, Gemini was by far the least frustrating model to observe.
-
comment
Comment #45878791
We have an example of a failed cross-tile result in the article - the models seem like they're much better at detecting whether something is in an image vs. identifying the boundar…
-
comment
Comment #45878741
Same! As we talk about in the article, the failures were less from raw model intelligence/ability than from challenges with timing and dynamic interfaces
-
comment
Comment #45878728
While running this I looked at hundreds and hundreds of captchas. And I still get rejected on like 20% of them when I do them. I truly don't understand their algorithm lol
-
comment
Comment #45878716
You could definitely do better than we do here - this was just a test of how well these general-purpose systems are out-of-the-box
- story
- story
-
comment
Comment #45096644
Roundtable | https://roundtable.ai | On-site San Francisco, CA | Full-time Roundtable is a research and deployment company building the proof-of-human layer in digital identity. Ro…
- story
- story
- story
-
comment
Comment #44379874
Co-founder of Roundtable here. I agree that better authentication methods for AI agents are needed. But right now bots and malicious agents are a real problem for anyone running si…
- story
-
comment
Comment #43165902
Seems like a lot of this could be explained by better food tending to be served in locations with lower commercial real estate prices (I believe Tyler Cowen has written about this)…
-
story
How we spot AI using keystrokes: Lessons from analyzing 5M+ survey responses
Each year, companies spend billions of dollars gathering survey data to guide product decisions. However, a growing percentage of this data is AI-generated. This bad data can lead …
- story
-
comment
Comment #41197766
Thanks for letting us know - we're planning to support TypeScript soon. A lot of stuff on our roadmap!