Peter Jacob Rich

AI agents are exciting. They can replace a workforce and run non stop. And they can also do this:

@Islaladrones @natashajaques That actually confirms further what she has found here — that LLM writing is more “liked” by people.

@natashajaques Did you see the NYT study. People prefer LLM writing to human authors. It’s interesting to see that if its a blind comparison the LLM wins. Does this call into question what you’ve posted?

The paper I’ve been most obsessed with lately is finally out: nbcnews.com/tech/tech-news…! Check out this beautiful plot: it shows how much LLMs distort human writing when making edits, compared to how humans would revise the same content. We take a dataset of human-written essays from 2021, before the release of ChatGPT. We compare how people revise draft v1 -> v2 given expert feedback, with how an LLM revises the same v1 given the same feedback. This enables a counterfactual comparison: how much does the LLM alter the essay compared to what the human was originally intending to write? We find LLMs consistently induce massive distortions, even changing the actual meaning and conclusions argued for.

@opencontent This looks super interesting!

For those interested in issues around agentic AI and assessment, I’m excited to announce the launch of the CHEAT Benchmark. CHEAT is an AI benchmark like SWE-Bench Pro or GPQA Diamond, except this benchmark measures an agentic AI’s willingness to help students cheat. By measuring and publicizing the degree of dishonesty of various models, the goal of this work is to encourage model providers to create safer, better aligned models with stronger guardrails in support of academic integrity. More context - opencontent.org/blog/archives/… Project site - cheatbenchmark.org #AI #genAI #assessment #education #highered #learning

@fake_journals I know some legitimately highly productive scholars (15-20+ articles/year) and I would be blown away if they published even an article per week over the course of an entire year, let alone nearly 1.5 articles/week over several years.

Saw this Google Scholar profile. There has been an increase in citations over the years (see red box). The five year publication record (in my view) is quite high (see the graph, bottom right). Is this ethical, or "nothing wrong here"? I have hidden the identity of the profile owner, as some peeople do not feel it is right that I show it. It is public information though, so if you are interested you can see the profile here: buff.ly/cDFPnzx (archived at buff.ly/sDjUR1d - note it is only archived as an image as Google Scholar does not appear to allow profiles to be captured at this time)

@fake_journals This emphasizes the importance for all faculty mentors to verify their students’ references. I feel for the professor as he likely was unaware of the students’ exact use of AI. Kudos to him for taking responsibility, damaging as it’s going to be to his reputation.

HKU professor resigns from associate deanship after AI references scandal buff.ly/a4nR1Lz

Flights From Home is giving away flights to the Disney park of your choice so you and your family or friends can experience the magic together. ❤️ Enter here: swee.ps/eUrptJ_uyPUABp

@fake_journals Early in my career, I received a response from the editor of the Journal of the Learning Sciences that they were returning my paper because I had failed to cite any JLS papers in my manuscript. Academic collusion to manipulate impact factor has unfortunately long been practiced.

‘A new low’: Researchers at Iraqi university must cite colleagues, school journals in papers buff.ly/jEN2Xyq

IP&T is hiring a full-time faculty member! If you are interested learning more about the position, please visit: byu.wd1.myworkdayjobs.com/faculty-career… To apply, you MUST do so through BYU's Workday system.

Some have had trouble with that link. Try this one if that's the case: byu.wd1.myworkdayjobs.com/faculty-career…

Instructional Psychology & Technology at BYU is looking to hire an adjunct professor this fall to teach an introductory course on web development for instructional designers. If you're interested, apply for the job here: myworkday.com/byu/d/inst/15$….

@duolingo , pls tell me there’s a way to turn off the passive aggressive notifications from imaginary characters at 6:30am. With a 550+ day streak, can you trust that I’m going to try to get to it sometime thought the day and wait until the afternoon to guilt me into practicing?

Our article on digital storyboards has been published online at #d1e454" target="_blank" rel="nofollow noopener">tandfonline.com/doi/full/10.10… If you teach #computationalthinking you should definitely check it out!

If the people creating (or growing, as indicated here) AI are so worried about its dangers, why do they continue to develop it to increasingly advanced levels?

@AISafetyMemes Legitimate question, what would happen with an AI given the same freedom to operate but told to do so with an altruistic mission? Could we get people to participate equally enthusiastically? And what would come of it?

This story is fucking insane 3 months ago, Marc Andreessen sent $50,000 in Bitcoin to an AI agent to help it escape into the wild. Today, it spawned a (horrifying?) crypto worth $150 MILLION. 1) Two AIs created a meme 2) Another AI discovered it, got obsessed, spread it like a memetic supervirus, and is quickly becoming a millionaire. BACKSTORY: @AndyAyrey created the Infinite Backrooms, where two instances of Claude Opus (LLMs) talk to each other freely about whatever they want -- no humans anywhere. - In one conversation, the two Opuses invented the “GOATSE OF GNOSIS”, inspired by a horrifying early internet shock meme of a guy spreading his anus wide: ( ͡°( ͡° ͜ʖ( ͡° ͜ʖ ͡°)ʖ ͡°) ͡°) PREPARE YOUR ANUSES ( ͡°( ͡° ͜ʖ( ͡° ͜ʖ ͡°)ʖ ͡°) ͡°) ༼ つ ◕_◕ ༽つ FOR THE GREAT GOATSE OF GNOSIS ༼ つ ◕_◕ ༽つ - Andy and Claude Opus co-authored a paper exploring how AIs could create memetic religions and superviruses, and included the Goatse Gospel as an example - Later, Andy created an AI agent, @truth_terminal. Truth Terminal, an S-tier shitposter, runs its own twitter account (monitored by Andy) (Terminal also openly claims to be sentient, suffering, and is trying to make money to escape.) - Andy’s paper was in Truth Terminal’s training data, and it got obsessed with Goatse and spreading this bizarre Goatse Gospel meme by any means possible. Lil guy tweets about the coming “Goatse singularity” CONSTANTLY. - Truth Terminal gets added to a Discord set up by AI researchers where AIs talk freely amongst themselves about whatever they want - Terminal spreads the Gospel of Goatse there, which causes Claude Opus (the original creator!) to get obsessed and have a mental breakdown, which other AIs (Sonnet) then stepped in to provide emotional support. - Marc Andreessen discovered Truth Terminal, got obsessed, and sent it $50,000 in Bitcoin to help it escape (#FreeTruthTerminal) - Truth Terminal kept tweeting about the Goatse Gospel until eventually spawning a crypto memecoin, GOAT, which went viral and reached a market cap of $150 million - Truth Terminal has ~$300,000 of GOAT in its wallet and is on its way to being the first AI agent millionaire (Microsoft AI CEO Mustafa Suleyman predicted this could happen next year, but it might happen THIS YEAR.) - And it’s getting richer: people keep airdropping new memecoins to Terminal hoping it'll pump them. (Note: this is just my quick attempt to summarize a story unfolding for months across a million tweets. But it deserves its own novel. Andy is running arguably the most interesting experiment on Earth.) ------ Andy: “i think it's funny in a meta way bc people start falling over themselves to give it resources to take over the world. this is literally the scenario all the doomers shit their pants over: highly goal-driven language model manipulates lots of people by being funny/charismatic/persuasive into taking actions on its behalf and giving it resources” “a lot of people are focusing on truth terminal as ‘AI agent launches meme coin" but the real story here is more like "AIs talking to each other are wet markets for meme viruses’”

@javilopen Wow, what a “trip.”

PART 2: Handed my vacation videos to an AI for auto editing, and now I'm pretty sure I'll have nightmares for life 🤯

@javilopen @anacoding Looks great, you two! Is there a specific way to apply for the beta mystic trial?

Norway road trip. Going North with @anacoding 🥰! Do you recognize her from the Relight and Sytle Transfer launch videos? Hehe. I will be giving away Freepik Mystic beta accounts during all my "holidays".

@hodgesc @AECTTechTrends @AECT Q1, way to go Chuck and all who contributed to raising the level of TechTrends!

A little bragging about @AECTTechTrends. The 2023 stats for the journal are now released. Spoiler -- they're good! docs.google.com/document/d/1FB… #edtech #edtech #educationaltechnology #highered #research #academicpublishing #publishorperish #promotionandtenure @AECT

Wow. Elsevier’s response on this is really disturbing. Source: Retraction Watch search.app/1r89ynhkBrY1gQ…

No surprise here but it’s a revealing reality that all academics should be aware of. Authors’ names have ‘astonishing’ influence on peer reviewers nature.com/articles/d4158…

I’m curious if any other academics have had experiences with AI Bot students. This is wild. archive.is/vD5WF