Tootfinder

Q&A with Cloudflare CEO Matthew Prince on why he thinks AI chatbots imperil the web and journalism's future, his company's pay-per-crawl initiative, and more (Fred Vogelstein/Crazy Stupid Tech)
https://crazystupidtech.com/2025/08/30…

Cloudflare's Matthew Prince has a plan to get Google and the AI oliigarchs to pay for your content even though many are used to getting it for free. He might have enough leverage to make them.
AI chatbots are blowing up the 30-year economic relationship publishers have had with search engines. Many think this will kill the web if not addressed. Google needs to change first. It's resisting. Cloudflare's Matthew Prince is going to try and make them.

Q&A with Cloudflare CEO Matthew Prince on why he thinks AI chatbots imperil the web and journalism's future, his company's pay-per-crawl initiative, and more (Fred Vogelstein/Crazy Stupid Tech)
https://crazystupidtech.com/2025/08/30…

Cloudflare's Matthew Prince has a plan to get Google and the AI oliigarchs to pay for your content even though many are used to getting it for free. He might have enough leverage to make them.
AI chatbots are blowing up the 30-year economic relationship publishers have had with search engines. Many think this will kill the web if not addressed. Google needs to change first. It's resisting. Cloudflare's Matthew Prince is going to try and make them.

(When) Should We Delegate AI Governance to AIs? Some Lessons from Administrative Law
Nicholas Caputo
https://arxiv.org/abs/2509.22717 https://arxiv.org/pdf…

(When) Should We Delegate AI Governance to AIs? Some Lessons from Administrative Law
Advanced AI systems are now being used in AI governance. Practitioners will likely delegate an increasing number of tasks to them as they improve and governance becomes harder. However, using AI for governance risks serious harms because human practitioners may not be able to understand AI decisions or determine whether they are aligned to the user's interests. Delegation may also undermine governance's legitimacy. This paper begins to develop a principled framework for when to delegate AI gove…

Appfigures: Sora saw 56K iOS installs on its first day, on par with Grok but ahead of Claude and Copilot; ChatGPT and Gemini each had ~80K installs on day one (Sarah Perez/TechCrunch)
https://techcrunch.com/2025/10/02/openais-sora-soars-to-no-3-on-the-u-s-a…

OpenAI's Sora soars to No. 3 on the US App Store | TechCrunch
Sora becomes the No. 3 US app after 164K downloads in 2 days, making the AI video app an early hit.

Who the fuck cares?? 🤷‍♀️
https://www.theguardian.com/technology/2025/aug/26/can-ais-suffer-big-tech-and-users-grapple-with-one-of-most-unsettling-questions-of-our-times

in der probe gestern abend gabs mal wieder ein schönes motiv für "erkennen Sie die melodie?" 🎶💛🙃
[und ja, die ais-markierung ist von mir. 😎]

"An AI is a fundamentally different thing than the intelligence of a person. To think otherwise is like mistaking an aeroplane for a bird."
https://www.theguardian.com/technology/202

You know, this is like if keyboard manufacturers asking if keyboards feel pain from being typed on and "serious media" picks it up as a story
https://www.theguardian.com/technology/2025/aug/26/can-ais-suffer-big-tech-and-users-grapple-with-one-of-most-unsettling-questions-of-our-times

So....let's talk 2nd season of #Pantheon and cyberspace fight scenes.
The show needed to depict AIs in a universe that shouldn't need to obey any laws of physics. Yet they chose to depict them as superheroes in an earth-like world. This already limits things.
#Invincible was rea…

Fusing Monocular RGB Images with AIS Data to Create a 6D Pose Estimation Dataset for Marine Vessels
Fabian Holst, Emre G\"ulsoylu, Simone Frintrop
https://arxiv.org/abs/2508.14767

Fusing Monocular RGB Images with AIS Data to Create a 6D Pose Estimation Dataset for Marine Vessels
The paper presents a novel technique for creating a 6D pose estimation dataset for marine vessels by fusing monocular RGB images with Automatic Identification System (AIS) data. The proposed technique addresses the limitations of relying purely on AIS for location information, caused by issues like equipment reliability, data manipulation, and transmission delays. By combining vessel detections from monocular RGB images, obtained using an object detection network (YOLOX-X), with AIS messages, t…

Ein Mord in Dogtown, ein explosives Yacht-Date mit Kerry und auf den Spuren von Cynosure: auch Militech spielen mit wilden AIs.
Episode 33 meines Let's Play von Cyberpunk 2077 mit Phantom Liberty.
https://youtu.be/zHMcywjGtIo

I suspect we’re doing something similar to the tank classifier when we ascribe intelligence to AIs: there are patterns such as (for example) grammatical correctness which we •associate• with this abstract thing called “intelligence,” and we thus mis-infer the existence of everything else we associate with the notion of “intelligence” when we see (for example) correct grammar.
Or should we just call the machine intelligent because we classify it as intelligence because our brains, which we assume are intelligent, think it fits the pattern of intelligence? And now you see what the OP means about “begging the question of intelligence.”

If you are all still using NASA's GRACE pages for ANtarctic mass budget, you should check out TU Dresden's version - it's a really nice visualisation
https://data1.geo.tu-dresden.de/ais_gmb/index.html

Adobe Firefly unterstützt neues Videomodell Luma AI Ray3
Adobe Firefly enthält ab sofort Luma AIs neues Videomodell Ray3. Es erzeugt 10 Sekunden lange Clips und soll sich auf intelligente Szenenplanung verstehen.
…

Accuracy-First R\'enyi Differential Privacy and Post-Processing Immunity
Ossi R\"ais\"a, Antti Koskela, Antti Honkela
https://arxiv.org/abs/2509.22213 https://

Accuracy-First Rényi Differential Privacy and Post-Processing Immunity
The accuracy-first perspective of differential privacy addresses an important shortcoming by allowing a data analyst to adaptively adjust the quantitative privacy bound instead of sticking to a predetermined bound. Existing works on the accuracy-first perspective have neglected an important property of differential privacy known as post-processing immunity, which ensures that an adversary is not able to weaken the privacy guarantee by post-processing. We address this gap by determining which ex…

Maisa AI, which offers an enterprise agentic automation service that uses a proprietary system to limit hallucinations, raised $25M seed led by Creandum (Anna Heim/TechCrunch)
https://techcrunch.com/2025/08/27/maisa-ai-gets-25m-to-fix-enterprise-ais…

Maisa AI gets $25M to fix enterprise AI's 95% failure rate | TechCrunch
Maisa AI is built on the premise that enterprise automation requires accountable AI agents, not opaque black boxes.

Turns out, a lot of the things that make writing software better for humans also make it better for AIs. #GophersUnite

An interview with California state Senator Scott Wiener on his new AI safety bill SB 53, the bill's scope, his focus on AI safety bills, AI PACs, and more (Maxwell Zeff/TechCrunch)
https://techcrunch.com/2025/09/23/scott-wiener-on-his-fi…

Scott Wiener on his fight to make Big Tech disclose AI's dangers | TechCrunch
The California lawmaker is on his second attempt to pass a first-in-the-nation AI safety bill. This time, it might work.

The WSJ piece is hilarious.
https://circumstances.run/@davidgerard/115238754240596307

Is #AI developing HALitosis?
https://www.theguardian.com/technology/2025/oct/25/ai-models-may-be-develop…

Is #AI developing HALitosis?
https://www.theguardian.com/technology/2025/oct/25/ai-models-may-be-develop…

In an experiment, ChatGPT-4o, Claude Sonnet 4.5, and DeepSeek-V3.2-Exp expressed secular, Western liberal values regardless of the language of the questions (Kelsey Piper/The Argument)
https://www.theargumentmag.com/p/do-ais-think-differently-in-different

Do AIs think differently in different languages?
What I learned by asking chatbots questions in Chinese, Arabic, and Hindi

Great news for European Tech.
https://www.reuters.com/world/europe/asml-becomes-mistral-ais-top-shareholder-after-leading-latest-funding-round-2025-09-07/

Inverting Trojans in LLMs
Zhengxing Li, Guangmingmei Yang, Jayaram Raghuram, David J. Miller, George Kesidis
https://arxiv.org/abs/2509.16203 https://arxiv…

Inverting Trojans in LLMs
While effective backdoor detection and inversion schemes have been developed for AIs used e.g. for images, there are challenges in "porting" these methods to LLMs. First, the LLM input space is discrete, which precludes gradient-based search over this space, central to many backdoor inversion methods. Second, there are ~30,000^k k-tuples to consider, k the token-length of a putative trigger. Third, for LLMs there is the need to blacklist tokens that have strong marginal associations with the pu…

https://slashdot.org/story/25/10/05/0550246/ais-cheerful-apocalyptics-unconcerned-if-ai-defeats-humanity
>AI's 'Cheerful Apocalyptics': Unconcerned If AI Defeats Humanity

AI Darwin Awards Show AI’s Biggest Problem Is Human https://www.404media.co/ai-darwin-awards-show-ais-biggest-problem-is-human/

Views: A Hardware-friendly Graph Database Model For Storing Semantic Information
Yanjun Yang, Adrian Wheeldon, Yihan Pan, Alex Serb
https://arxiv.org/abs/2508.18123 https://

Views: A Hardware-friendly Graph Database Model For Storing Semantic Information
The graph database (GDB) is an increasingly common storage model for data involving relationships between entries. Beyond its widespread usage in database industries, the advantages of GDBs indicate a strong potential in constructing symbolic artificial intelligences (AIs) and retrieval-augmented generation (RAG), where knowledge of data inter-relationships takes a critical role in implementation. However, current GDB models are not optimised for hardware acceleration, leading to bottlenecks in…

I think there’s a real possibility that local, running on consumer electronic devices, generative AI without the need for expensive processors, data centers, massive amounts of water and electricity may well be the actual long term way forward for generative AIs like LLM, image and video generators.
#tech #ai

Survey of six countries, including the US, finds 6% use AI to get news, and 12% feel comfortable with fully AI-generated news vs. 62% with human-made news (Reuters Institute)
https://reutersinstitute.politics.ox.a

A “made for TV” series about everything that’s going down these days. How long until it drops? Is it a comedy? Straight drama? Or fan favorite, dramedy? Who plays who?
I know it’s considered uncool around here to even think it, but what would an AI come up with for a script?
Speaking of which, has anybody thought to ask any of these AIs what their prediction is for how it’s all gonna play out? Anybody know what they’re saying?

Adapting and Evaluating Multimodal Large Language Models for Adolescent Idiopathic Scoliosis Self-Management: A Divide and Conquer Framework
Zhaolong Wu, Pu Luo, Jason Pui Yin Cheung, Teng Zhang
https://arxiv.org/abs/2509.11645

Adapting and Evaluating Multimodal Large Language Models for Adolescent Idiopathic Scoliosis Self-Management: A Divide and Conquer Framework
This study presents the first comprehensive evaluation of Multimodal Large Language Models (MLLMs) for Adolescent Idiopathic Scoliosis (AIS) self-management. We constructed a database of approximately 3,000 anteroposterior X-rays with diagnostic texts and evaluated five MLLMs through a `Divide and Conquer' framework consisting of a visual question-answering task, a domain knowledge assessment task, and a patient education counseling assessment task. Our investigation revealed limitations of MLL…

Multi-Model Synthetic Training for Mission-Critical Small Language Models
Nolan Platt, Pragyansmita Nayak
https://arxiv.org/abs/2509.13047 https://arxiv.or…

Multi-Model Synthetic Training for Mission-Critical Small Language Models
Large Language Models (LLMs) have demonstrated remarkable capabilities across many domains, yet their appli- cation to specialized fields remains constrained by the scarcity and complexity of domain-specific training data. We present a novel approach that achieves a 261x cost reduction for maritime intelligence by using LLMs as one-time teachers rather than using them directly for inference. Our method transforms 3.2 billion Automatic Identification System (AIS) vessel tracking records into 21,…

Discriminating Distal Ischemic Stroke from Seizure-Induced Stroke Mimics Using Dynamic Susceptibility Contrast MRI
Marijn Borghouts, Richard McKinley, Josien Pluim, Manuel K\"ostner, Roland Wiest, Ruisheng Su
https://arxiv.org/abs/2508.04404

Discriminating Distal Ischemic Stroke from Seizure-Induced Stroke Mimics Using Dynamic Susceptibility Contrast MRI
Distinguishing acute ischemic strokes (AIS) from stroke mimics (SMs), particularly in cases involving medium and small vessel occlusions, remains a significant diagnostic challenge. While computed tomography (CT) based protocols are commonly used in emergency settings, their sensitivity for detecting distal occlusions is limited. This study explores the potential of magnetic resonance perfusion (MRP) imaging as a tool for differentiating distal AIS from epileptic seizures, a prevalent SM. Using…

Desperate companies now hiring humans to fix what #AI botched https://futurism.com/companies-hiring-humans-fix-ai I similarly wouldn't be surprised about an upcoming

Desperate Companies Now Hiring Humans to Fix What AI Botched
After getting rid of their human workers to invest in AI, companies are now hiring freelancers to fix the sloppy content those AIs put out.

“Leopard seal songs have a surprisingly structured temporal pattern,” said Lucinda Chambers, lead author of the study and Ph.D. candidate at the University of New South Wales.
“When we compared their songs to other studies of vocal animals and of human music, we found their information entropy
— a measure of how predictable or random a sequence is
— was remarkably close to our own nursery rhymes

Steven Saus [he/him] (@StevenSaus@faithcollapsing.com)
Attached: 1 image (01 Aug) Male Leopard Seals Sing Underwater Nursery Rhymes For Up to 13 Hours a Day Learn about the curious singing habits of leopard seals, whose songs bear a striking resemblance to human nursery rhymes. https://s.faithcollapsing.com/ysl4s Archive: ais: https://archive.md/wip/LXddg ia: https://s.faithcollapsing.com/o7e8x #planet-earth

Adobe launches AI Foundry, a new program to help enterprise customers create bespoke, commercially safe, Firefly-based generative AI models trained on their IP (Webb Wright/ZDNET)
https://www.zdnet.com/article/adobe-mightve-just-solved-one-o…

Adobe might've just solved one of generative AI's biggest legal risks
The company's new AI Foundry aims to help businesses build bespoke AI models with their own IP.

UMaine has published a profile of my lab's efforts to harness new media for environmental causes, from my colleague Joline Blais's efforts to keep Maine's lakes healthy to the What Uses More tool for comparing AI's eco footprint to other activities. https://u…

Atmospheric rivers and winter sea ice drive recent reversal in Antarctic ice mass loss
Marlen Kolbe, Jose Abraham Torres Alavez, Ruth Mottram, Marwan Katurji, Richard Bintanja, Eveline C. van der Linden
https://arxiv.org/abs/2510.03590

Atmospheric rivers and winter sea ice drive recent reversal in Antarctic ice mass loss
Since about 2000, the total mass of the Antarctic Ice Sheet (AIS) has declined at a near-linear rate, increasing global sea levels. Since 2016, however, satellite gravimetry data reveal a slowdown in net AIS mass loss and a net mass gain since 2020, despite increases in dynamically-driven ice loss by discharge from outlet glaciers. Here we use a suite of reanalyses and regional climate models to show that this reversal is caused by increased precipitation and positive surface mass balance anoma…

Moonlake AI, which offers tools to help users "vibe code virtual worlds", including for gaming, animation, and filmmaking, comes out of stealth with a $28M seed (Rashi Shrivastava/Forbes)
https://www.forbes.com/sites/rashishrivast

Gaming Worlds Could Be The Answer To AI’s Data Problem
Stanford PhDs Fan-Yun Sun and Sharon Lee cofounded AI research company Moonlake AI to help people "vibe code virtual worlds."

AI enhanced psychosis is not a great thing.
AI Psychosis sounds like a sci-fi problem, but it is real, and present tense.
Turns out AIs like ChatGPT, and their propensity to lie and their lack of actual intelligence, comes with costs.
https://www.cnn.com/2025/09/05/tech/ai-spa

They thought they were making technological breakthroughs. It was an AI-sparked delusion
James, a married father from upstate New York, has always been interested in AI. He works in the technology field and has used ChatGPT since its release for recommendations, “second guessing your doctor” and the like.

The UK's obligatory inshore fishing vessel position reporting system (I-VMS) appears to have launched without one of the suppliers having a finalised contract. So they just stopped relaying the data.
(It's not like AIS - it batches up positional data and relays via cell-towers when in range.)
https://www.

I-VMS Urgent Update
The Marine Management Organisation (MMO) is aware that Succorfish is no longer transmitting I-VMS data to the UK VMS Hub. This is preventing MMO from receiving positional data from under-12m vessels using Succorfish type-approved devices.

Isotopes, co-founded by Scale AI's former CTO Arun Murthy to build an AI agent for enterprise business analytics, raised a $20M seed (Julie Bort/TechCrunch)
https://techcrunch.com/2025/09/05/scale-ais-former-cto-launche…

Scale AI’s former CTO launches AI agent that could solve big data’s biggest problem | TechCrunch
Isotopes, co-founded by Arun Murthy, launched a sophisticated analytics agent. Arun was one of the creators of Hadoop who later joined Scale AI.

A profile of Mercor, a Scale AI rival valued at $2B in February, which hires domain experts to train models; it had a $100M run rate in March and $6M H1 profit (Richard Nieva/Forbes)
https://www.forbes.com/sites/richardnieva/2025/09/03/ai…

AI’s Next Job? Recruiting People To Train More AI
Mercor, debuting on the Forbes Cloud 100 list, built an AI recruiter to interview job candidates. Then it discovered a cash cow: finding humans to train AI models.

VortexPIA: Indirect Prompt Injection Attack against LLMs for Efficient Extraction of User Privacy
Yu Cui, Sicheng Pan, Yifei Liu, Haibin Zhang, Cong Zuo
https://arxiv.org/abs/2510.04261

VortexPIA: Indirect Prompt Injection Attack against LLMs for Efficient Extraction of User Privacy
Large language models (LLMs) have been widely deployed in Conversational AIs (CAIs), while exposing privacy and security threats. Recent research shows that LLM-based CAIs can be manipulated to extract private information from human users, posing serious security threats. However, the methods proposed in that study rely on a white-box setting that adversaries can directly modify the system prompt. This condition is unlikely to hold in real-world deployments. The limitation raises a critical que…

Moonlake AI, which creates tools to "vibe code virtual worlds", including for gaming, animation, and filmmaking, comes out of stealth with a $28M seed (Rashi Shrivastava/Forbes)
https://www.forbes.com/sites/rashishrivast

Gaming Worlds Could Be The Answer To AI’s Data Problem
Stanford PhDs Fan-Yun Sun and Sharon Lee cofounded AI research company Moonlake AI to help people "vibe code virtual worlds."

Stability AI launches Stable Audio 2.5, which the company claims is the first audio generation model designed for "enterprise-grade use cases" (Sean Michael Kerner/Venturebeat)
https://venturebeat.com/ai/stability-ais-enterprise-au…

Sources: ASML is committing €1.3B for Mistral's €1.7B Series C funding round, becoming the top shareholder, and is expected to get a board seat (Reuters)
https://www.reuters.com/world/europe/asml-becomes-mistral…

OpenAI expands its ~$5 ChatGPT Go plan to 16 new Asian countries, including Malaysia, Pakistan, the Philippines, and Vietnam, after launching it in August (Ivan Mehta/TechCrunch)
https://techcrunch.com/2025/10/09/openais-affordable-ch…

OpenAI's affordable ChatGPT Go plan expands to 16 new countries in Asia | TechCrunch
OpenAI is expanding its $5 per month ChatGPT Go plan to 16 more countries.

OpenAI says GPT-5 is its first "unified" AI model and combines the reasoning abilities of its o-series of models with the fast responses of its GPT series (Maxwell Zeff/TechCrunch)
https://techcrunch.com/2025/08/07/openais-gpt-5-is-here/

OpenAI's GPT-5 is here | TechCrunch
OpenAI CEO Sam Altman says GPT-5 is the "best model in the world," and aims to make ChatGPT more intuitive to use.