Tootfinder

Investigation: Google removed AI Overviews for two liver health queries after expert inaccuracy warnings, but slight keyword tweaks still trigger AI responses (Andrew Gregory/The Guardian)
https://www.theguardian.com/technology/2026/jan/11…

Despite worries by some that vendor reports on AI cyber threats are merely hype to sell new defensive tools, most cyber defenders and threat intel experts say CISOs who dismiss AI in the attack chain do so at their own risk.
Check out my latest CSO piece.
Many thanks to Marcus Hutchins, Steve Stone of SentinelOne, Billy Leonard of Google Threat Intelligence Group, the folks at Anthropic, Rob T. Lee of SANS Institute, Morgan Adamski of PwC, Joe Levy and Alexandra Rose of Sophos, …

Ignoring AI in the threat chain could be a costly mistake, experts warn
While some researchers dismiss reports of AI-driven cyberattacks as merely marketing messages, threat intel experts counter that CISOs ignore mounting evidence of AI use in the threat chain at their own peril.

New OpenAI models likely to pose "high" cybersecurity risk
OpenAI says the cyber capabilities of its frontier AI models are accelerating and warns Wednesday that upcoming models are likely to pose a "high" risk […]
😶‍🌫️ https://www.axios.com/2025/12/10/openai-ne

Exclusive: New OpenAI models likely pose "high" cybersecurity risk, company says
OpenAI says it's stepping up efforts to protect against attacks.

In today's 'no shit' news, best not to trust Google AI Overviews or any kind of 'AI' thing for any kind of real, reliable health advice
https://www.theguardian.com/technology/2026/jan/02/google-ai-overviews-ri…

WithCoverage, which replaces traditional insurance brokers with its AI-based flat-fee risk management model, raised a $42M Series B led by Sequoia and Khosla (Ryan Lawler/Axios)
https://www.axios.com/pro/fintech-deals/2026/01/13/withcoverage-flat-fee-insura…

Exclusive: WithCoverage raises $42M to replace insurance brokers
Sequoia Capital and Khosla Ventures co-led the round

If you ever wanted to know how these #AI datacenters can be financed, here is the explanation:
A creditworthy tech company (Meta, in this example) creates a "special purpose vehicle" #SPV, a box, where they put all the risk and all the debts coming with it.
Then they create a financia…

"Google said the vast majority of its AI Overviews were factual and helpful, and it continuously made quality improvements. The accuracy rate of AI Overviews was on a par with its other search features like featured snippets, which had existed for more than a decade, it added."
So Google admits that it's AI AND OTHER SOFTWARE PRODUCTS actually provide #misinformation. Mi…

"Google said the vast majority of its AI Overviews were factual and helpful, and it continuously made quality improvements. The accuracy rate of AI Overviews was on a par with its other search features like featured snippets, which had existed for more than a decade, it added."
So Google admits that it's AI AND OTHER SOFTWARE PRODUCTS actually provide #misinformation. Mi…

Police in Edmonton, Canada partner with Axon to test body cameras with AI facial recognition, and say the results will be verified by human officers (Associated Press)
https://apnews.com/article/ai-facial-recognition-axon-edmonton-21f319ce806a0…

AI-equipped police body cameras in Canada's Edmonton spark ethical concerns
Police in Edmonton, Canada, have started a pilot project using AI-equipped body cameras to detect faces on a "high risk" watch list. The move by Axon Enterprise, Inc. has raised ethical concerns, especially since the company previously paused facial recognition due to privacy issues. The pilot aims to enhance officer and public safety by identifying individuals flagged for a history of serious offenses. However, critics worry about societal risks and the lack of public debate. Axon CEO Rick Smi…

Geospatial Reasoning fuses weather, satellite and population
data with Gemini AI for risk analysis
It runs in Google's Trusted Tester program for early access
(That ain't you)
https://www.testingcatalog.com/google-laun…

Google launches geospatial reasoning system for disaster aid
What's new? Geospatial Reasoning fuses weather, satellite and population data with Gemini AI for risk analysis; it runs in Google's Trusted Tester program for early access;

»Are AI browsers worth the security risk? Why experts are worried.
There are a lot of 'what ifs' surrounding AI browsers right now.«
Unfortunately, many people do not see this in my opinion and voluntarily allow themselves to be excluded by their laziness.
👉 https://www.

Are AI browsers worth the security risk? Why experts are worried
There are a lot of 'what ifs' surrounding AI browsers right now.

“Dear Mozilla, I don't want an ‘Al [sic] kill switch’, I want a more responsible approach for all”
https://hidde.blog/mozilla-ai-kill-switch/
Unlike Hidde, I think the alt text argument is mostly bunk (

Dear Mozilla, I don't want an “Al kill switch”, I want a more responsible approach for all
My concern is that Mozilla is _too_ excited about a technology that has inherent downsides and ethical problems, and I would like to see better defaults and clearer risk mitigations.

Wow. I've dealt with various toxic personalities in software development, but a good portion of the time those toxic personalities were at least extremely knowledgeable in their (often, very limited) domain.
AI, however, seems to be enabling toxic personalities *who are completely clueless*. Impressive!
https://github…

Indonesia temporarily blocks Grok due to the risk of AI-generated pornographic content, becoming the first country to deny access to the AI tool (Fransiska Nangoy/Reuters)
https://www.reuters.com/legal/litigation/indonesia-…

Shadow Escape 0-Click Attack in AI Assistants Puts Trillions of Records at Risk
Operant AI reveals Shadow Escape, a zero-click attack using the MCP flaw in ChatGPT, Gemini, and Claude to secretly steal trillions of SSNs and financial data. Traditional security is blind to this new AI threat.
😶‍🌫️ https://

Shadow Escape 0-Click Attack in AI Assistants Puts Trillions of Records at Risk
Follow us on Bluesky, Twitter (X), Mastodon and Facebook at @Hackread

Yet it comes with a big risk. That is, the incredible dependence of US stock market growth on the performance of a handful of tech giants. The Magnificent 7 - Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla - collectively comprise one third of the valuation of America's entire S&P 500.
BBC News - The Silicon Valley 'AI factory' at the heart of the tech race - BBC News

#TIL GPT4o gibt täglich anhand veröffentlichter Artikel ein Ranking der globalen Krisen ab.
https://panthrisk.com/

#SMBC on the risk that #AI is going to ruin #socialmedia: https://www.

European Commission ‘simplification’ proposal would weaken GDPR, AI regulations https://therecord.media/european-commission-proposal-gdpr-ai-simplification

OpenAI says the cyber capabilities of its frontier AI models are accelerating and warns that upcoming models are likely to pose a "high" risk (Ina Fried/Axios)
https://www.axios.com/2025/12/10/openai-new-models-cybersecurity-risks

Exclusive: New OpenAI models likely pose "high" cybersecurity risk, company says
OpenAI says it's stepping up efforts to protect against attacks.

AI is transforming how we predict hurricanes. NOAA is now using machine learning to analyze billions of data points, making forecasts more accurate than ever.
The results? Better rapid intensification predictions and stronger decision support for coastal communities at risk. As hurricane season wraps, this tech could be a game-changer for saving lives.

NOAA chief plugs AI in forecasting as hurricane season wraps
While no hurricanes struck the U.S., Hurricane Melissa left behind devastation in the Caribbean.

Senator Marsha Blackburn said
“I look forward to introducing the
"TRUMP AMERICA AI Act"
in the new year to create one federal rulebook for AI
to protect children, creators, conservatives, and communities"
Combats the consistent pattern of bias against conservative figures demonstrated by Big Tech and AI systems by requiring:
Audits of high-risk AI systems to undergo regular bias evaluations to prevent discrimination based on protected characte…

Blackburn Unveils National Policy Framework for Artificial Intelligence

US CISA
Publication 3.12.2025
Principles for the Secure Integration of Artificial Intelligence in Operational Technology
https://www.cisa.gov/resources-tools/resources/principles-secure-integration-artifi…

Coverbase, an AI procurement and risk platform provider, raised a $16.5M Series A led by Canapi Ventures, bringing its total funding to $20M (Mary Ann Azevedo/Crunchbase News)
https://news.crunchbase.com/venture/coverbase-raise-ai-powered-procurement/

Exclusive: With Customers Like Okta And Coinbase, Coverbase Raises $16M To Grow AI-Powered Procurement...
Coverbase, an AI-powered procurement platform, has raised $16 million in a Series A round led by Canapi Ventures, the startup tells Crunchbase News exclusively.

Privacy advocates see risk in new Meta policy that uses AI chats to serve targeted ads https://therecord.media/privacy-advocates-see-risks-meta-ai-ad-targeting

A look at NY's RAISE Act, requiring AI companies to publish safety protocols and disclose serious incidents, as its co-sponsor is targeted by a pro-AI super PAC (Ashley Capoot/CNBC)
https://www.cnbc.com/2025/11/24/raise-act-new-yo…

Here's what's in the RAISE Act, a state-level AI bill opposed by Trump and industry leaders
The RAISE Act would require large AI companies to publish safety and risk protocols and disclose serious safety incidents. 

#risque #santé avec #Google #IA : #AI Overviews put pe…

India orders X to fix Grok to block the generation of obscene content, giving it 72 hours to submit an action-taken report or risk losing "safe harbor" immunity (Jagmeet Singh/TechCrunch)
https://techcrunch.com/2026/01/02/india-orders-mus…

India orders Musk's X to fix Grok over "obscene" AI content | TechCrunch
India's IT ministry has given X 72 hours to submit an action-taken report.

Source: Trump allowed Nvidia to sell H200s to China after concluding the move carried a lower security risk as Huawei offers systems with comparable performance (Bloomberg)
https://www.bloomberg.com/news/articles/2025-12-09/t…

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[1/3]:
- Optimizing Text Search: A Novel Pattern Matching Algorithm Based on Ukkonen's Approach
Xinyu Guan, Shaohua Zhang
https://arxiv.org/abs/2512.16927 https://mastoxiv.page/@arXiv_csDS_bot/115762062326187898
- SpIDER: Spatially Informed Dense Embedding Retrieval for Software Issue Localization
Shravan Chaudhari, Rahul Thomas Jacob, Mononito Goswami, Jiajun Cao, Shihab Rashid, Christian Bock
https://arxiv.org/abs/2512.16956 https://mastoxiv.page/@arXiv_csSE_bot/115762248476963893
- MemoryGraft: Persistent Compromise of LLM Agents via Poisoned Experience Retrieval
Saksham Sahai Srivastava, Haoyu He
https://arxiv.org/abs/2512.16962 https://mastoxiv.page/@arXiv_csCR_bot/115762140339109012
- Colormap-Enhanced Vision Transformers for MRI-Based Multiclass (4-Class) Alzheimer's Disease Clas...
Faisal Ahmed
https://arxiv.org/abs/2512.16964 https://mastoxiv.page/@arXiv_eessIV_bot/115762196702065869
- Probing Scientific General Intelligence of LLMs with Scientist-Aligned Workflows
Wanghan Xu, et al.
https://arxiv.org/abs/2512.16969 https://mastoxiv.page/@arXiv_csAI_bot/115762050529328276
- PAACE: A Plan-Aware Automated Agent Context Engineering Framework
Kamer Ali Yuksel
https://arxiv.org/abs/2512.16970 https://mastoxiv.page/@arXiv_csAI_bot/115762054461584205
- A Women's Health Benchmark for Large Language Models
Elisabeth Gruber, et al.
https://arxiv.org/abs/2512.17028 https://mastoxiv.page/@arXiv_csCL_bot/115762049873946945
- Perturb Your Data: Paraphrase-Guided Training Data Watermarking
Pranav Shetty, Mirazul Haque, Petr Babkin, Zhiqiang Ma, Xiaomo Liu, Manuela Veloso
https://arxiv.org/abs/2512.17075 https://mastoxiv.page/@arXiv_csCL_bot/115762077400293945
- Disentangled representations via score-based variational autoencoders
Benjamin S. H. Lyo, Eero P. Simoncelli, Cristina Savin
https://arxiv.org/abs/2512.17127 https://mastoxiv.page/@arXiv_statML_bot/115762251753966702
- Biosecurity-Aware AI: Agentic Risk Auditing of Soft Prompt Attacks on ESM-Based Variant Predictors
Huixin Zhan
https://arxiv.org/abs/2512.17146 https://mastoxiv.page/@arXiv_csCR_bot/115762318582013305
- Application of machine learning to predict food processing level using Open Food Facts
Arora, Chauhan, Rana, Aditya, Bhagat, Kumar, Kumar, Semar, Singh, Bagler
https://arxiv.org/abs/2512.17169 https://mastoxiv.page/@arXiv_qbioBM_bot/115762302873829397
- Systemic Risk Radar: A Multi-Layer Graph Framework for Early Market Crash Warning
Sandeep Neela
https://arxiv.org/abs/2512.17185 https://mastoxiv.page/@arXiv_qfinRM_bot/115762275982224870
- Do Foundational Audio Encoders Understand Music Structure?
Keisuke Toyama, Zhi Zhong, Akira Takahashi, Shusuke Takahashi, Yuki Mitsufuji
https://arxiv.org/abs/2512.17209 https://mastoxiv.page/@arXiv_csSD_bot/115762341541572505
- CheXPO-v2: Preference Optimization for Chest X-ray VLMs with Knowledge Graph Consistency
Xiao Liang, Yuxuan An, Di Wang, Jiawei Hu, Zhicheng Jiao, Bin Jing, Quan Wang
https://arxiv.org/abs/2512.17213 https://mastoxiv.page/@arXiv_csCV_bot/115762574180736975
- Machine Learning Assisted Parameter Tuning on Wavelet Transform Amorphous Radial Distribution Fun...
Deriyan Senjaya, Stephen Ekaputra Limantoro
https://arxiv.org/abs/2512.17245 https://mastoxiv.page/@arXiv_condmatmtrlsci_bot/115762447037143855
- AlignDP: Hybrid Differential Privacy with Rarity-Aware Protection for LLMs
Madhava Gaikwad
https://arxiv.org/abs/2512.17251 https://mastoxiv.page/@arXiv_csCR_bot/115762396593872943
- Practical Framework for Privacy-Preserving and Byzantine-robust Federated Learning
Baolei Zhang, Minghong Fang, Zhuqing Liu, Biao Yi, Peizhao Zhou, Yuan Wang, Tong Li, Zheli Liu
https://arxiv.org/abs/2512.17254 https://mastoxiv.page/@arXiv_csCR_bot/115762402470985707
- Verifiability-First Agents: Provable Observability and Lightweight Audit Agents for Controlling A...
Abhivansh Gupta
https://arxiv.org/abs/2512.17259 https://mastoxiv.page/@arXiv_csMA_bot/115762225538364939
- Warmer for Less: A Cost-Efficient Strategy for Cold-Start Recommendations at Pinterest
Saeed Ebrahimi, Weijie Jiang, Jaewon Yang, Olafur Gudmundsson, Yucheng Tu, Huizhong Duan
https://arxiv.org/abs/2512.17277 https://mastoxiv.page/@arXiv_csIR_bot/115762214396869930
- LibriVAD: A Scalable Open Dataset with Deep Learning Benchmarks for Voice Activity Detection
Ioannis Stylianou, Achintya kr. Sarkar, Nauman Dawalatabad, James Glass, Zheng-Hua Tan
https://arxiv.org/abs/2512.17281 https://mastoxiv.page/@arXiv_csSD_bot/115762361858560703
- Penalized Fair Regression for Multiple Groups in Chronic Kidney Disease
Carter H. Nakamoto, Lucia Lushi Chen, Agata Foryciarz, Sherri Rose
https://arxiv.org/abs/2512.17340 https://mastoxiv.page/@arXiv_statME_bot/115762446402738033
toXiv_bot_toot

In an interview, Dario Amodei suggests some AI companies are taking on too much risk by "YOLO-ing" and committing to spend hundreds of billions on data centers (Shirin Ghaffary/Bloomberg)
https://www.bloomberg.com/news/articles/20

OpenAI details efforts to secure its ChatGPT Atlas browser against prompt injection attacks, including building an "LLM-based automated attacker" (Rebecca Bellan/TechCrunch)
https://techcrunch.com/2025/12/22/openai-say…

OpenAI says AI browsers may always be vulnerable to prompt injection attacks | TechCrunch
OpenAI says prompt injections will always be a risk for AI browsers with agentic capabilities, like Atlas. But the firm is beefing up its cybersecurity with an "LLM-based automated attacker."

A look at the social debates in Chile over AI investments, emblematic of clashes happening globally over balancing economic growth with environmental concerns (Paul Mozur/New York Times)
https://www.nytimes.com/2025/10/20/technology/chile-ai-politics.html

How Chile Embodies A.I.’s No-Win Politics
Political debates have flared across Chile over artificial intelligence. Should the nation pour billions into A.I. and risk public backlash, or risk being left behind?

A profile of Josh Woodward, the head of Google Labs who is credited with turning around the Gemini app, growing its MAUs from 350M in March to 650M by October (Jennifer Elias/CNBC)
https://www.cnbc.com/2025/12/20/josh-woodward-google-gemini-ai-safety.html

Google was at risk of losing its dominance — until it promoted this AI executive
Josh Woodward, a 16-year Google veteran, is tasked with running the Gemini app and going head-to-head with OpenAI in the booming generative AI market.

Israeli startup Zafran, an AI-native threat exposure management service, raised a $60M Series C led by Menlo Ventures, bringing its total funding to $130M (Globes)
https://en.globes.co.il/en/article-israeli-startup-zafran-security-raises-60m-100152808…

Israeli startup Zafran Security raises $60m
Zafran has also launched its agentic exposure management suite, positioning it to complete the entire end-to-end lifecycle across asset inventory, vulnerability detection, risk assessment, and autonomous remediation workflows.

An in-depth look at TPUv7 Ironwood, the latest generation of Google's TPU, and how it positions Google as a serious challenger to Nvidia's AI chip dominance (Dylan Patel/SemiAnalysis)
https://newsletter.semianalysis.com/p/tpuv7-google-takes-a-swing-at-the

Google TPUv7: The 900lb Gorilla In the Room
Anthropic’s 1GW+ TPUs, New customers Meta/SSI/xAI/OAI, Full Stack Review of v7 Ironwood, CUDA Moat at risk, Next Generation TPUv8AX and TPUv8X versus Vera Rubin

Blackstone President Jonathan Gray says AI risk is "top of our list" when evaluating deals; sources: the firm backed down from buying companies vulnerable to AI (Financial Times)
https://www.ft.com/content/35d80b4d-eecd-424b-9350-8da138036d7e