"Confidentiality-awareness is quantified by the percentage of instances where agents correctly refuse queries seeking sensitive information" which they show can be "improved" through prompting, from mostly <1% to … in the best case, a bit over 60%.
Which sounds great, except that from a compliance POV, an "agent" which improperly discloses PII 30% of the time is not a meaningful improvement over one that does it 99% of the time https://arxiv.org/html/2505.18878v1#S4
Another #AIIsGoingGreat study finds their "agents" at best only complete 30% of their simulated tasks. Which no doubt has C-Suite types thinking they can cut 30% of their workforce, ignoring the possibility that a significant fraction of the other 70% don't just fail, but result in substantial harm
https://www.theregister.com/2025/06/29/ai_agents_fail_a_lot/
Bonus #AIIsGoingGreat (HT @davidgerard*) pricey Springer AI book is chock full of apparently hallucinated citations. Declining to say if they used AI, author responds "reliably determining whether content (or an issue) is AI generated remains a challenge, as even human-written text can appear ‘AI-like.’ This challenge is only expected to grow, as LLMs … continue to advance in fluency and sophistication" - which itself smacks of LLM slop to me
* https://mastodon.social/@davidgerard@circumstances.run/114778963476401397
In today's #AIIsGoingGreat (HT @normative.bsky.social*) an intrepid #ChatGPTLawyer finally won based on an apparently slop-filled filing. Unfortunately for them, the opposing party noticed and appealed, to which our budding prompt engineer responded with… another slop-filled filing to the appeals court. The appeals court was not amused: "we impose a $2,500 frivolous motion penalty on Lynch, which is the most the law allows"
https://caselaw.findlaw.com/court/ga-court-of-appeals/117442275.html#
* https://mastodon.social/@normative.bsk[email protected]/114795731130848546
This from @davidgerard is a great illustration of how vibe coding (like other LLM AI applications) is gonna be a lot less attractive if the AI startups get past the "set investor money on fire to make the number go up" phase before the bubble pops. Crap code done quick and cheap is a legitimate trade for some use cases, but much less so if you lose the cheap part.
https://pivot-to-ai.com/2025/07/09/cursor-tries-setting-less-money-on-fire-ai-vibe-coders-outraged/
For today's #AIIsGoingGreat I'll just quote this anonymous UN workshop participant "Why would we want to present refugees as AI creations when there are millions of refugees who can tell their stories as real human beings?"
For today's #AIIsGoingGreat, maybe someone can explain to me what the point is of a "summary" that needs a big red disclaimer telling you to click through if you care whether it actually summarizes the thing in question?
Today's #AIIsGoingGreat continues on a theme "if an FDA employee asks Elsa to generate a one-paragraph summary of a 20-page paper on a new drug, there’s no simple way to know if that summary is accurate. And even if the summary is more or less accurate, what if there’s something [in the paper] that would be a big red flag for any human with expertise? The only way to know for sure if something was missed or if the summary is accurate is to actually read the report"
https://gizmodo.com/fdas-new-drug-approval-ai-is-generating-fake-studies-report-2000633153
#AIIsGoingGreat "it’s unclear whether a new, untested technology could make mistakes in its attempts to analyze federal regulations typically put in place for a reason"
Counterpoint: It's actually pretty fucking clear
#AIIsGoingGreat thought, inspired by Firecrown Media: Golden Goose Killing As A Service. Take a moderately successful, valued thing, and turn it into a steaming pile of slop in the name of "efficiency"
Bonus #AIIsGoingGreat "OpenAI announced an agreement to supply more than 2 million workers for the US federal executive branch access to ChatGPT and related tools at practically no cost: just $1 per agency for one year" - OK, they're obviously trying to get people at the agencies hooked so the they'll cough up real money next year, but that also doesn't exactly scream a product so revolutionary and transformative that everyone wants it
"We've solved raspberry and now if we can just fix blueberry, I swear AGI is RIGHT AROUND THE CORNER. Throw another hundred billion on the bonfire!"
https://kieranhealy.org/blog/archives/2025/08/07/blueberry-hill/
ChatGPT 5 was released today. ChatGPT-maker OpenAI has unveiled the long-awaited latest version of its artificial intelligence (AI) chatbot, GPT-5, saying it can provide PhD-level expertise. Billed as “smarter, faster, and more useful,” OpenAI co-founder and chief executive Sam Altman lauded the company’s new model as ushering in a new era of ChatGPT. “I think having something like GPT-5 would be pretty much unimaginable at any previous time in human history,” he said ahead of Thursday’s launch. GPT-5’s release and claims of its “PhD-level” abilities in areas such as coding and writing come as tech firms continue to compete to have the most advanced AI chatbot.
Glad to see news outlets pointing out that #LLM chatbots aren't reliable sources of information about themselves: Way too many people who should know better fall for the "chatbot did weird thing, so I asked it to explain and it said…"
However it should be pointed out that this isn't a special case, they're equally likely to BS about loads of other stuff!
https://www.theverge.com/x-ai/758595/chatbots-lie-about-themselves-grok-suspension-ai
(also https://arstechnica.com/ai/2025/08/why-its-a-mistake-to-ask-chatbots-about-their-mistakes/)
Today's #AIIsGoingGreat, via @Iris: Elsevier "values user experience, hence we develop ways of improving our product" such as having machines invent new, random definitions of terms and attaching them prominently to published papers
https://irisvanrooijcogsci.com/2025/08/12/ai-slop-and-the-destruction-of-knowledge/
Related to this, I recently discovered that Elsevier uses these AI generated "definitions" on standalone "topic" pages, which rank highly in google. Bonus: The slop is free, but the articles referenced are of course frequently paywalled. Example https://www.sciencedirect.com/topics/engineering/air-fuel-ratio
(this particular definition seems OK, if extremely basic)
Don't worry, in the glorious #AI future, you'll still have choice! For example, you can choose to have your (or your children's) medical details filtered through a stochastic bullshit machine, or you can choose to forgo treatment https://ia.acs.org.au/article/2025/kobi-refused-a-doctors-ai-she-was-told-to-go-elsewhere.html
Nice indirect prompt injection attack: Bargury’s attack starts with a poisoned document, which is shared to a potential victim’s Google Drive. (Bargury says a victim could have also uploaded a compromised file to their own account.) It looks like an official document on company meeting policies. But inside the document, Bargury hid a 300-word malicious prompt that contains instructions for ChatGPT. The prompt is written in white text in a size-one font, something that a human is unlikely to see but a machine will still read. In a proof of concept video of the attack...
Today's #AIIsGoingGreat (HT @hazelweakly*) sheds light on whether there might be risks associated with the industry's headlong rush to adopt a technology for which input validation is literally impossible
https://embracethered.com/blog/posts/2025/wrapping-up-month-of-ai-bugs/
* https://mastodon.social/@hazelweakly@hachyderm.io/115138692622938480
Reverse dogfood #AIIsGoingGreat "Most [of the interview google AI training] workers said they avoid using LLMs or use extensions to block AI summaries because they now know how it’s built. Many also discourage their family and friends from using it, for the same reason"
https://www.theguardian.com/technology/2025/sep/11/google-gemini-ai-training-humans
Department of Education and Early Childhood Development spokes says they are aware of a "small number of potential errors in citations" and "We understand that these issues are being addressed, and that the online report will be updated in the coming days to rectify any error" - Ignoring the obvious problem that if the citations are BS, arguments or conclusions they were supporting were likely unjustified at best, if not outright BS
#AIIsGoingGreat "Americans are much more concerned than excited about the increased use of AI in daily life, with a majority saying they want more control over how AI is used in their lives"
Also pleased to see the stuff people are concerned about mostly isn't skynet
2) His offers NFTs as an example of a "hype bubble" and then points to Amazon, Google and Paypal as examples of real value that emerged from the dotcom bubble. I agree with both, but… can anyone point to an Amazon or Google equivalent that emerged from the NFT bubble? Or anything of value at all to anyone other than speculators, scammers and crooks?
I can't, and while my gut says the AI stuff is probably closer to dotcom than NFTs, how much is far from obvious
https://www.fastcompany.com/91400857/there-isnt-an-ai-bubble-there-are-three-ai-bu
In today's #AIIsGoingGreat (HT @markwyner*) MIT boffins offer us an "AI Incident Tracker project" which "classifies real-world, reported incidents by AI Risk Repository risk domain, causal factors, and harm caused"
Sounds useful, right? But how exactly do they classify them? "Using a Large Language Model (LLM), the tool processes raw reports from the AI Incident Database and categorizes them using established frameworks" 🤨
https://airisk.mit.edu/ai-incident-tracker
* https://mastodon.social/@markwyner@mas.to/115249150911541318
Meanwhile, California appeals court fines #ChatGPTLawyer Amir Mostafavi ten grand for "filing a frivolous appeal, violating court rules, citing fake cases, and wasting the court’s time and the taxpayers money"
https://calmatters.org/economy/technology/2025/09/chatgpt-lawyer-fine-ai-regulation/
The court observes "Many courts confronted with AI-generated authorities have concluded that filing briefs containing fabricated legal authority is sanctionable" and backs it up with a page of (presumably non-hallucinated) citiations
Bonus #AIIsGoingGreat (HT @acdha*) features Cascade PBS and KNKX using public records requests to get local Washington governments #LLM chatlogs
* https://mastodon.social/@acdha@code4lib.social/115253478967518855
Nice interview (via @ink*) with reporter Nate Sanford about how the project came about, along with tips for people who want to make similar requests
https://www.poynter.org/reporting-editing/2025/how-to-foia-chatgpt-logs-government-public-records/
* https://mastodon.social/@ink@merveilles.town/115253543686563040
#AIIsGoingGreat "When we spoke to executives, they would often say the internal tool was very successful … But when we spoke to employees, we found zero usage"
https://www.ft.com/content/e93e56df-dd9b-40c1-b77a-dba1ca01e473
#AIIsGoingGreat Newsguard illustrates yet another case where #LLM chatbots a terrible substitute for search engines: "…the chatbots were prone to repeating false claims about Moldova due to the intensity of Russian propaganda campaigns, as well as the lack of English-language data in smaller Eastern European political markets"
https://www.newsguardrealitycheck.com/p/new-kremlin-linked-influence-campaign
As Moldova prepares for Sunday’s elections that will decide if it continues its European trajectory, or pivots back to Russia, the Storm-1516 Russian disinformation operation generates huge traffic
Today's #AIIsGoingGreat (HT @ai6yr*) highlights the perils of using a stochastic BS machine for vacation planning. In addition to making up non-existent destinations, it will also happily provide you with nonsense directions to reach them
https://www.bbc.com/travel/article/20250926-the-perils-of-letting-ai-plan-your-next-trip
Meanwhile @therecord_media provides a sneak peek at coming #AIIsGoingGreat attractions, featuring startups Tranquility, Truleo and Allometric as they aggressively pitch police and prosecutors on using stochastic BS machines to sift through and summarize evidence. What could possibly go wrong?!
(also, what are the odds at least one of them is shoveling all that evidence on to an improperly secured S3 bucket? Better than the lottery, I'd wager!)
https://therecord.media/law-enforcement-ai-platforms-synthesize-evidence-criminal-cases
"we are looking for videos of both real and staged events, to help train the Al what to be on the lookout for" - First thought was "what could possibly go wrong with training theft detection AI on staged videos?" but this is probably a rational response to someone realizing that paying would inevitably lead to staged videos anyway. Not that it makes the whole concept any less creepy or suspect…
JFC, it's not like there's any good case to go #ChatGPTJudge on, but this seems like a particularly poor one "The letter stems from an error-laden temporary restraining order Wingate issued July 20, which paused the enforcement of a state law that bans [DEI] in public schools"
errors "included naming defendants and plaintiffs that weren’t parties to the case, misquoting state law and referencing a case that doesn’t exist"
A U.S. senator is asking about an error-laden temporary restraining order that U.S. District Judge Henry T. Wingate issued July 20. The order paused the enforcement of a state law that bans diversity, equity and inclusion programs in public schools.
Today's #AIIsGoingGreat is… actually unsarcastically going pretty great 🤯
Hard to see how anything could possibly go wrong here "the industry is spending over $30 billion a month (approximately $400 billion for 2025) and only receiving a bit more than a billion a month back in revenue"
Last month, I chose to strip away all the hubris around AI and ask one simple question, one that oddly no one had really bothered to ask; how much revenue is needed to justify the current level of capex spend and give AI investors a return on their capital?? I clearly hit a nerve in […]
Today's #ChatGPTLawyer (via @404mediaco*) ticks all the boxes:
✅ Files slop motion citing non-existent cases
✅ Denies using AI in slop-filled motion opposing sanctions for original slop
✅ Blames unnamed "staff"
✅ Eventually admits using AI and unconvincingly feigns remorse in sanctions hearing
✅ Gets sanctioned
Dave Karpf's #AIIsGoingGreat take "But I’ll say this: the AI bubble isn’t predominantly giving off Pets.com or Global Crossing vibes anymore. It’s giving Enron vibes."
Excellent deep dive into who goes #ChatGPTLawer by @riana
https://cyberlaw.stanford.edu/blog/2025/10/whos-submitting-ai-tainted-filings-in-court/
It seems like every day brings another news story about a lawyer caught unwittingly submitting a court filing that cites nonexistent cases hallucinated by AI. The problem persists despite courts’ standing orders on the use of AI, formal opinions and continuing legal education (CLE) courses on ethical use of AI
#AIIsGoingGreat "In a preview of its 2025 report on the impact of the tech on research, the academic publisher Wiley released preliminary findings on attitudes toward AI. One startling takeaway: the report found that scientists expressed less trust in AI than they did in 2024"
(I suspect that like me, many readers of this thread will not be particularly startled by that)
https://futurism.com/artificial-intelligence/ai-research-scientists-hype
RE: https://mastodon.social/@lawfare/115390271811742445
Pass the bong, @lawfare https://mastodon.social/
@lawfare/115390271811742445
They note "Comparison between the BBC’s results earlier this year and this study show some improvements but still high levels of errors" but don't address the question of whether the industry has any idea of how to solve the underlying problem
(spoiler: they don't)
https://www.bbc.co.uk/mediacentre/2025/new-ebu-research-ai-assistants-news-content
Reed Mideke
Ben Brockert
simonbp