Dwayne Ripley (@dwayneripley) header
Dwayne Ripley (@dwayneripley) profile picture

No longer using this tainted platform. Find me on Bluesky.

Sydney, New South Wales
Joined March 2016
Photo by Dwayne Ripley (@dwayneripley)Photo by Dwayne Ripley (@dwayneripley)Photo by Dwayne Ripley (@dwayneripley)Photo by Dwayne Ripley (@dwayneripley)Photo by Dwayne Ripley (@dwayneripley)Photo by Dwayne Ripley (@dwayneripley)Photo by Dwayne Ripley (@dwayneripley)Photo by Dwayne Ripley (@dwayneripley)Photo by Dwayne Ripley (@dwayneripley)
I was fortunate enough to contribute as an author and editor to this recently published book alongside some wonderful academics here in Sydney and in Denmark. If you are interested in educational design know-how, it open access. :) taylorfrancis.com/books/edit…
2
3
130
Dwayne Ripley retweeted
LLM hallucination is just math fitting a smooth curve and producing the most likely continuation. It’s matrix multiplication, so it will always output something that maps to a language object. None of this has anything to do with consciousness. Whenever humans invent a new tech - from fire to writing to steam engines - we project god onto it and use it to explain things we don’t yet understand - for eg consciousness.
Replying to @protosphinx
Let’s assume your stream of consciousness with its random thoughts is the same like an LLM hallucinating Why not simply add a value function which filters the thoughts which are in line with your identity definition and store important points ins the memory And voila . You hab conciousness 😎
70
182
10
2,182
90,629
Dwayne Ripley retweeted
i have no more passwords left in me
146
2,845
107
17,934
309,719
Dwayne Ripley retweeted
Christopher Alexander has an observation about problem solving that I like: you should always be focusing on solving the part that has the fewest degrees of freedom. When figuring out how to design a kitchen, for instance, there are a bunch of subproblems to solve: where to put the stove and the windows and the kitchen table. And which of these have the fewest degrees of freedom? The windows. If you want good light, there is going to be only one wall where you can place the windows, and at best two spots on that wall where the window looks natural. So you put the window there. And now what? The kitchen table, because you want to have that where the good light falls. The stove can wait because that can sit nearly anywhere. If you start by placing the stove, there is a big risk that you block the only good position for one of the other subproblems that have fewer degrees of freedom, and so the whole design will suffer.
54
272
39
3,420
188,003
Dwayne Ripley retweeted
I don’t want Chat GPT. I want free education. I don’t want AI. I want clean air and coral reefs.
200
11,109
39
54,965
438,374
Dwayne Ripley retweeted
There are a LOT more papers that need to be desk rejected at ICLR. Somehow, this hallucination wasn't caught. So, I went on a crazy rabbit hole and found 50 more (many are just as funny). We're approaching crisis levels Paper titles included below 👇 gptzero.me/news/iclr-2026/
This paper has been desk rejected. LLM-generated papers that hallucinate references and do not report LLM usage will be desk rejected per ICLR policy (blog.iclr.cc/2025/08/26/poli…) Reviewers of other versions of this submission have been notified.
19
46
23
401
179,150
Dwayne Ripley retweeted
I wrote about the deluge of AI slop that is drowning scientific journals and preprint servers theatlantic.com/science/2026…
10
60
6
153
7,126
Dwayne Ripley retweeted
Public debates are often framed around: One study. One lawsuit. One headline. But reliable knowledge doesn’t come from isolated data points. It comes from: Replication. Dose and exposure. All available evidence. We need better ways of explaining how conclusions are reached.
12
23
1
151
2,944
Dwayne Ripley retweeted
Incredible. One of the world’s most prestigious conferences, NeurIPS 2025, has published over 50 papers containing fake citations, likely hallucinated by generative AI. Several of these papers report affiliations from top-tier institutions, including NYU, Cornell, and Princeton. And this is likely only the tip of the iceberg. If citations are hallucinated, what other parts of the papers are hallucinated as well? Generative AI has the potential to destroy the peer review system and, consequently, trust in published research. The cost of producing a paper is decreasing, while the cost of reviewing remains constant, or is even increasing. This is unsustainable. We really need to radically rethink the review system.
9
31
5
108
9,861
America is officially a Jerry Springer episode
115
372
34
4,056
41,858
Dwayne Ripley retweeted
For the sake of Aaron Swartz I beg you to make this shit stop.
68
394
37
2,798
171,830
Dwayne Ripley retweeted
Accountants are saying they are spending 4–10 HOURS A WEEK cleaning up AI slop. WOW.
78
298
36
4,779
98,266
When you use AI to cut corners in academic publishing, you disrespect editors' and reviewers' time and endanger their reputations. You will be judged and it will be justified. Your aim should not be to get published. It should be to produce something worthy of publication.
1
16
Dwayne Ripley retweeted
Depression is a sign you’re paying attention. Anxiety is a sign you need to stop.
172
1,701
108
12,498
382,333
Dwayne Ripley retweeted
A new article on how knowledge is central to students developing transformational accounts of their undergraduate education. It is not instrumentality that's the problem in education but the absence of a focus on knowledge. link.springer.com/article/10…
1
27
3
53
4,067
Dwayne Ripley retweeted
Don't let what you see on social media skew your view of humanity. Most bad behavior is due to a few bad apples: • Just 3% of active users are toxic, but they generate 33% of online content • 1% of communities launch 74% of conflict • 0.1% of users spread 80% of fake news @jayvanbavel podcasts.apple.com/us/podcas…
88
951
145
2,662
286,367
A beautiful day today for a visit to the University of Sydney Book fair. It's an experience. sydney.edu.au/engage/visit/c…
2
1
3
127