Tootfinder

Meta auf Einkaufstour: Weitere KI-Experten wechseln den Arbeitgeber
Apple, OpenAI, Anthropic und Safe Superintelligence verlieren KI-Experten. Meta baut sein KI-Team konsequent weiter aus.
http…

Super Bowl champion launches LGBTQ football camp, aims to create unifying safe space within sports

https://www.cbssports.com/nfl/news/super-b

The AI Safe Harbor Moratorium Undermines Mrs. Palsgraf’s Remedy
https://musictechpolicy.com/2025/05/28/the-ai-safe-harbor-moratorium-undermines-mr…

The AI Safe Harbor Moratorium Undermines Mrs. Palsgraf’s Remedy
The AI Moratorium is a rent seeker on a free ride.

Ilya Sutskever becomes the CEO of Safe Superintelligence, the AI startup he co-founded after leaving OpenAI; Meta reportedly poached previous CEO Daniel Gross (Ashley Capoot/CNBC)
https://www.cnbc.com/2025/07/03/ilya-sutsk

It's like a 3D puzzle. This 18th-century Italian safe has the most unique unlocking sequences ever.

"How can you make sure your sunscreen is 'reef safe'?"
#Suncream #Environment
https:/…

Eigenlijk was dat best Eem lekker besturingssysteem... 😀
"Aangezien XP geen backdoors heeft mbt persoonlijke data wordt het gezien als safe."
En als de Russen die Amerikaanse software veilig genoeg vinden voor zichzelf, zal dat dus ook wel kloppen.
#Microsoft #Rusland

I'm going to give sudo-rs a go on all my personal Linux machines and every other one that is Debian 13 (trixie) onwards. I shall report back if this makes my life hard.
https://github.com/trifectatechfoundation/sudo-rs

Ban all new fossil fuel projects for a safe future! #environment

We don’t need fossil fuels to power Europe. - Greenpeace International
Fossil gas, oil, and coal are not just driving the climate crisis – they’re making Europe dependent on bullies like Putin and Trump.

Woman in a construction company wants to wear a skirt over her work pants, which doesn't sound very safe.
Religion Clause: EEOC Sues Over Denial of Dress Code Religious Accommodation for Apostolic Christian Employee
http://religionclause.blogspot.com/2025/06/eeoc-sues-over-denial-of-dress-code.html

lowkey wish i had an eye in my hand. but with better (on-by-default) natural cover than eyelids. it'd be an absolute bitch to keep safe the way it's always depicted in art

Safe. Everywhere. Always. | A Canberra Community Survey
https://docs.google.com/forms/d/e/1FAIpQLSdSrRIhL9NhacbREQUPvZ-KN7VN15VelwkoXwIHFCDLq4qgHg/viewform

Safe. Everywhere. Always. | A Canberra Community Survey
Women should be safe no matter where they are, what they’re doing or what they’re wearing. In response to recent horrific events of men’s violence against women in our community, Senator for the ACT and Minister for Women Katy Gallagher is inviting all Canberra women and girls to share their experiences and ideas for change in the survey below. A summary of key findings from the survey will be made public following its conclusion, alongside a list of commitments and proposed next steps. A…

15 #rust #cli tools that will make you abandon bash scripts forever
https://

These #LA #ICE protests and pushback are so hopeful. Send the kidnappers home empty-handed.
"We keep us safe" in action.

🇺🇦 #NowPlaying on KEXP's #Audioasis
Day Soul Exquisite:
🎵 NSH (Not Safe Here)
#DaySoulExquisite

This https://arxiv.org/abs/2406.02436 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csRO_…

Safe, Out-of-Distribution-Adaptive MPC with Conformalized Neural Network Ensembles
We present SODA-MPC, a Safe, Out-of-Distribution-Adaptive Model Predictive Control algorithm, which uses an ensemble of learned models for prediction, with a runtime monitor to flag unreliable out-of-distribution (OOD) predictions. When an OOD situation is detected, SODA-MPC triggers a safe fallback control strategy based on reachability, yielding a control framework that achieves the high performance of learning-based models while preserving the safety of reachability-based control. We demonst…

A Representation Engineering Perspective on the Effectiveness of Multi-Turn Jailbreaks
Blake Bullwinkel, Mark Russinovich, Ahmed Salem, Santiago Zanella-Beguelin, Daniel Jones, Giorgio Severi, Eugenia Kim, Keegan Hines, Amanda Minnich, Yonatan Zunger, Ram Shankar Siva Kumar
https://arxiv.org/abs/2507.02956

A Representation Engineering Perspective on the Effectiveness of Multi-Turn Jailbreaks
Recent research has demonstrated that state-of-the-art LLMs and defenses remain susceptible to multi-turn jailbreak attacks. These attacks require only closed-box model access and are often easy to perform manually, posing a significant threat to the safe and secure deployment of LLM-based systems. We study the effectiveness of the Crescendo multi-turn jailbreak at the level of intermediate model representations and find that safety-aligned LMs often represent Crescendo responses as more benign…

🇺🇦 #NowPlaying on #BBC6Music's #CerysMatthews
Kutiman:
🎵 So Long Stay Safe
#Kutiman
https://kutiman.bandcamp.com/track/so-long-stay-safe
https://open.spotify.com/track/6pzWTmqXfxdzwzr8eXbdsc

If the goal truly is to
“make America healthy,”
why are HHS life-saving programs for babies and young children being eliminated?
Here are just two examples,
of small and inexpensive live-saving programs being cut or gutted.
⭐️At a time when sudden unexpected infant deaths are on the rise,
the National Institutes of Health (NIH) has quietly pulled funding for the "Safe to Sleep" campaign,
a program that cut deaths from Sudden Infant Death S…

We have booked a stay in a self-catering cottage in Gloucestershire for a week over the summer. We've just been sent the welcome pack and it reads a little bit like solving an old school adventure game.
There is a keysafe at the cottage - located on the left pillar of the porch when you walk up the path. The key safe code is XXXX.
Use the small key in here to open the wooden box on your left as you walk in to the porch. Inside that you will find a large key which is for the main door.
Please make sure you leave this key in the wooden box on your departure as this is the only key to the property.

Happy Pride Month! Wishing all who celebrate a safe and glamorous time.
GDEP is always a safe and supportive ally space for the LGBTIQA community.
#GDEPAU #EOLD #AskAnnetta

#kde kmymoney also started to eat data. If Devs turn something from reliable and safe to unreliable and unsafe, the real question of liability comes into play. How can such a thing end in distribution?
Maybe we need to re-organize open source to have a reliable portfolio of applications, a kind of reliable #kde fr…

LOL… I didn’t think it was possible and I guess Santa Ono didn’t either.
His hasty resignation from #UMich seemed like a safe move but it is never safe to depend on a public sector taken over by Fascists.
#Florida #Michigan

Board votes against Santa Ono as UF president amid conservative protest
Florida’s higher education board late Tuesday voted against confirming Santa Ono as the next president of the University of Florida after questioning …

This https://arxiv.org/abs/2506.03612 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDS_…

Connectivity-Preserving Minimum Separator in AT-free Graphs
Let $A$ and $B$ be disjoint, non-adjacent vertex-sets in an undirected, connected graph $G$, whose vertices are associated with positive weights. We address the problem of identifying a minimum-weight subset of vertices $S\subseteq V(G)$ that, when removed, disconnects $A$ from $B$ while preserving the internal connectivity of both $A$ and $B$. We call such a subset of vertices a connectivity-preserving, or safe minimum $A,B$-separator. Deciding whether a safe $A,B$-separator exists is NP-hard …

Control Synthesis in Partially Observable Environments for Complex Perception-Related Objectives
Zetong Xuan, Yu Wang
https://arxiv.org/abs/2507.02942 http…

Control Synthesis in Partially Observable Environments for Complex Perception-Related Objectives
Perception-related tasks often arise in autonomous systems operating under partial observability. This work studies the problem of synthesizing optimal policies for complex perception-related objectives in environments modeled by partially observable Markov decision processes. To formally specify such objectives, we introduce \emph{co-safe linear inequality temporal logic} (sc-iLTL), which can define complex tasks that are formed by the logical concatenation of atomic propositions as linear ine…

Freedom means having your family home safe and sound. This 4th of July, we’re demanding an end to the abductions to keep our families whole.
#FreeAmerica #EndtheAbductions

Don't read the comments.
Think you have some free time and nothing better to do: Do anything, just
DONT READ THE COMMENTS
Mastodon is not safe from that.
You are not protected on the fediverse from reading the comments.
A classic trap is the "use linux now, it's far superior than any stupid OS you are using RIGHT NOW!!!"
vs
"Pardon, couldn't help but overhear, I love my WIndows and these poor mega corps need it for their w…

Situation: There are 13 different bike, transit, and safe streets advocacy groups in SF.
“13?! This is ridiculous! We should make a single group that covers everyone’s priorities!”
Situation: There are 14 different bike, transit, and safe streets advocacy groups in SF.

This https://arxiv.org/abs/2412.13800 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_hepp…

$e^+ e^- \to μ^+ μ^-$ in the Asymptotically Safe Standard Model
We study the electron-positron to muon--anti-muon cross-section in the asymptotically safe Standard Model. In particular, we include the graviton contributions to the scattering amplitude, which is computed from momentum-dependent timelike one-particle-irreducible correlation functions. Specifically, we employ reconstruction techniques for the graviton spectral functions. We find that the full asymptotically safe quantum cross section decreases in the ultraviolet with the centre-of-mass energy,…

A Practical SAFE-AI Framework for Small and Medium-Sized Enterprises Developing Medical Artificial Intelligence Ethics Policies
Ion Nemteanu, Adir Mancebo Jr., Leslie Joe, Ryan Lopez, Patricia Lopez, Warren Woodrich Pettine
https://arxiv.org/abs/2507.01304

A Practical SAFE-AI Framework for Small and Medium-Sized Enterprises Developing Medical Artificial Intelligence Ethics Policies
Artificial intelligence (AI) offers incredible possibilities for patient care, but raises significant ethical issues, such as the potential for bias. Powerful ethical frameworks exist to minimize these issues, but are often developed for academic or regulatory environments and tend to be comprehensive but overly prescriptive, making them difficult to operationalize within fast-paced, resource-constrained environments. We introduce the Scalable Agile Framework for Execution in AI (SAFE-AI) desig…

End-to-end learning of safe stimulation parameters for cortical neuroprosthetic vision https://www.biorxiv.org/content/10.1101/2025.01.23.634543v1.full "typical range of 10-20 μA", so special measures must be taken to prevent that a 10,000-electrode…

This https://arxiv.org/abs/2308.13140 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csRO_…

Learn With Imagination: Safe Set Guided State-wise Constrained Policy Optimization
Deep reinforcement learning (RL) excels in various control tasks, yet the absence of safety guarantees hampers its real-world applicability. In particular, explorations during learning usually results in safety violations, while the RL agent learns from those mistakes. On the other hand, safe control techniques ensure persistent safety satisfaction but demand strong priors on system dynamics, which is usually hard to obtain in practice. To address these problems, we present Safe Set Guided Stat…

They put these footprints on the grates in the sidewalks in downtown Calgary. I guess to show it’s safe to walk on them? I can’t say it’s on all the grates but I don’t remember seeing one without them.
Edit: I’m embarrassed I didn’t think of this on my own, but @… pointed out these are very helpful for people wearing heels!

Safe AF Podcast
Great Australian Pods Podcast Directory: #GreatAusPods

Is It Safe To Learn And Share? On Psychological Safety and Social Learning in (Agile) Communities of Practice
Christiaan Verwijs, Evelien Acun-Roos, Daniel Russo
https://arxiv.org/abs/2507.01065

Is It Safe To Learn And Share? On Psychological Safety and Social Learning in (Agile) Communities of Practice
As hybrid, distributed, and asynchronous work models become more prevalent, continuous learning in Agile Software Development (ASD) gains renewed importance. Communities of Practice (CoPs) are increasingly adopted to support social learning beyond formal education, often relying on virtual communication. Psychological safety, a prerequisite for effective learning, remains insufficiently understood in these settings. This mixed-methods study investigates psychological safety within Agile CoPs th…

Lucky summer moon
Observing what we endure
From a safe distance.
#dailyhaikuprompt - summer moon
#haiku
#poem

🇺🇦 #NowPlaying on #BBC6Music's #Glastonbury
Wolf Alice:
🎵 Safe From Heartbreak (Glastonbury 2025)
#WolfAlice
https://testwolfalice.bandcamp.com/track/safe-from-heartbreak-if-you-never-fall-in-love
https://open.spotify.com/track/2bR9noN9BRUrgEcatelUvb

This https://arxiv.org/abs/2503.09117 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

GRU: Mitigating the Trade-off between Unlearning and Retention for Large Language Models
Large language model (LLM) unlearning has demonstrated its essential role in removing privacy and copyright-related responses, crucial for their legal and safe applications. However, the pursuit of complete unlearning often comes with substantial costs due to its compromises in their general functionality, leading to a notorious trade-off between unlearning and retention. In examining the update process for unlearning dynamically, we find gradients hold essential information for revealing this …

Isn't it time that the Palestinians are given a safe place to live, to escape potential annihilation? A country to call their own? There are some practically empty areas in the USA – think Wyoming, for instance. Wouldn't that be an idea?

Design Didn't Prevent These Hazard Exposures
David. E Mertz (Fermilab)
https://arxiv.org/abs/2506.03280 https://arxiv.org/pdf/250…

Design Didn't Prevent These Hazard Exposures
Two recent electrical incidents demonstrate how the design of equipment encouraged electrical workers to take actions that violated NFPA 70E principles. The features that encouraged non-compliant work execution will be described, as well as how the equipment was improved to facilitate safe work practices. Design-stage processes that help identify features that will foster rather than compromise safe work practices will be identified as well.

And just 30 years ago, #PFAS were generally viewed as safe. Of course, it appears the manufacturers knew otherwise (https://en.wikipedia.org/wiki/PFAS), but hey, it made a good profit. Yet another case of…

And just 30 years ago, #PFAS were generally viewed as safe. Of course, it appears the manufacturers knew otherwise (https://en.wikipedia.org/wiki/PFAS), but hey, it made a good profit. Yet another case of…

"Safe and easy ways to recycle electricals – and even get money for your old gadgets"
#Electronics #Technology #Recycling

And so it begins.
https://blog.google/around-the-globe/google-europe/we-are-announcing-sparkasse-as-our-first-national-credential-partner-for-eu-age-assurance/

We are announcing Sparkasse as our first national credential partner for EU age assurance.
The lack of a reliable standard for safe, effective age checks across sites and apps online has long frustrated parents and companies. Google’s Credential Manager API — backed by Google Wallet and zero-knowledge proof cryptography — has proven that it’s possible. What’s needed now is collaboration and initiative to bring about the future of digital identity.That begins now. At the Global Digital Collaboration Conference today in Geneva, Germany’s Sparkasse, a trusted network of 343 …

This https://arxiv.org/abs/2501.04250 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

Publish on Ping: A Better Way to Publish Reservations in Memory Reclamation for Concurrent Data Structures
Safe memory reclamation techniques that utilize per read reservations, such as hazard pointers, often cause significant overhead in traversals of linked concurrent data structures. This is primarily due to the need to announce a reservation, and fence to enforce appropriate ordering, before each read. In read-intensive workloads, this overhead is amplified because, even if relatively little memory reclamation actually occurs, the full overhead of reserving records is still incurred while traver…

Daniel Gross is leaving Safe Superintelligence, the AI startup he co-founded with Ilya Sutskever, to join Meta's new superintelligence lab (Bloomberg)
https://www.bloomberg.com/news/articles/2025-07-03/meta-hi…

@… @…
https://mstdn.ca/@CycleWR/…

CycleWR (@CycleWR@mstdn.ca)
Attached: 1 image Tuesday is World Bicycle Day! It’s also day one of the Vélo Canada Bikes annual pedal poll 😀🚴🏼‍♀️ There’s still time to claim your volunteer count shift to help collect crucial cycling data. The poll runs all across Canada. We're focussed on Waterloo Region, but there's a good chance it's happening where you live. We need more safe places to ride and this is how we prove it. Sign up for a shift & join the movement! #Cycling #PedalPollSondoVelo #Waterlo…

🇺🇦 #NowPlaying on #BBC6Music's #Glastonbury
Wolf Alice:
🎵 Safe From Heartbreak (Glastonbury 2025)
#WolfAlice
https://testwolfalice.bandcamp.com/track/safe-from-heartbreak-if-you-never-fall-in-love
https://open.spotify.com/track/2bR9noN9BRUrgEcatelUvb

How to configure Magento 2 to send emails via AWS SES.
https://winkelwagen.de/2025/06/03/magento-2-use-amazon-ses/

Magento 2: Use Amazon SES
This only works on Magento 2.4+ You want to use Amazon AWS Simple Email Service to send your transaction mails? It took a moment, but maybe you want to safe the struggle, let me tell you, what to c…

"If your reports don't feel safe, they won't tell you" — This is one of the clearest and most important pieces of advice I've heard for managers.
It's a perfect illustration of the "monster in the middle dilemma for navigating both social and organizational/authoritative power dynamics as a manager. Power dynamics are the monster in the middle — and if a manager doesn't actively work to mitigate that, they will fail to operate effectively as a manager…

This https://arxiv.org/abs/2501.02620 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

Back to Base: Towards Hands-Off Learning via Safe Resets with Reach-Avoid Safety Filters
Designing controllers that accomplish tasks while guaranteeing safety constraints remains a significant challenge. We often want an agent to perform well in a nominal task, such as environment exploration, while ensuring it can avoid unsafe states and return to a desired target by a specific time. In particular we are motivated by the setting of safe, efficient, hands-off training for reinforcement learning in the real world. By enabling a robot to safely and autonomously reset to a desired reg…

Predictive Risk Analysis and Safe Trajectory Planning for Intelligent and Connected Vehicles
Zeyu Han, Mengchi Cai, Chaoyi Chen, Qingwen Meng, Guangwei Wang, Ying Liu, Qing Xu, Jianqiang Wang, Keqiang Li
https://arxiv.org/abs/2506.23999

Predictive Risk Analysis and Safe Trajectory Planning for Intelligent and Connected Vehicles
The safe trajectory planning of intelligent and connected vehicles is a key component in autonomous driving technology. Modeling the environment risk information by field is a promising and effective approach for safe trajectory planning. However, existing risk assessment theories only analyze the risk by current information, ignoring future prediction. This paper proposes a predictive risk analysis and safe trajectory planning framework for intelligent and connected vehicles. This framework fi…

I always find the use of first person plural pronouns in discussions of the distant future to be too cute…
Humans didn’t exist a million years ago. There were some crafty hominid apes but not the sort you could clean up and mistake for a human.
There is no “us” 7 billion years from now. There is very likely no “us” in half a million years. We may or may not have descendants but they won’t be “us” in any sense.
A billion years is longer than Earth has had visible life

Is it safe to use toilets in your dreams?

Hawaiian Airlines flights operating safely after cyberattack affects some IT systems https://therecord.media/hawaiian-airlines-cyberattack-flights-safe

Oh—EMERGENCY DEPARTMENT! That makes much more sense now. 🤦🏻‍♂️
#NeedMoreCoffee

Thread and Memory-Safe Programming with CLASS
Lu\'is Caires (Instituto Superior T\'ecnico)
https://arxiv.org/abs/2505.20848 https://

Thread and Memory-Safe Programming with CLASS
CLASS is a proof-of-concept general purpose linear programming language, flexibly supporting realistic concurrent programming idioms, and featuring an expressive linear type system ensuring that programs (1) never misuse or leak stateful resources or memory, (2) never deadlock, and (3) always terminate. The design of CLASS and the strong static guarantees of its type system originates in its Linear Logic and proposition-as-types foundations. However, instead of focusing on its theoretical found…

Safe bet that on any given afternoon, 25-50% of the people who teach intro classes at Brown CS are out riding a bike. Sometimes we even run into each other! And since it's Andy van Dam that people really want to see, not me, here's a shot of just him w/ his new bike.

🇺🇦 #NowPlaying on #BBC6Music's #CraigCharles
Massive Attack:
🎵 Safe From Harm
#MassiveAttack
https://refel.bandcamp.com/track/massive-attack-safe-from-harm-refel-t-remix
https://open.spotify.com/track/1vNNfTgHsrpOXeiaXQBlH7

🔊 #NowPlaying on #BBCRadio3:
#UltimateCalm
- Hania Rani: Series 4
Embrace transformation with Hania Rani and some peaceful morphing musical selections that gently ebb and flow. Plus, the Danish artist Agnes Obel shares her musical safe havens.
Relisten now 👇
https://www.bbc.co.uk/programmes/m002bhw9

Global ocean acidification has passed safe planetary boundary threshold: Study https://news.mongabay.com/short-article/2025/06/global-ocean-acidification-has-passed-safe-planetary-boundary-threshold-study/

Global ocean acidification has passed safe planetary boundary threshold: Study
A new assessment finds that the world’s oceans crossed the safe threshold for acidification in 2020, breaching a key planetary boundary and posing serious threats to marine life. Ocean acidification is caused when excess atmospheric carbon dioxide, resulting from human activities like burning fossil fuels, dissolves in seawater, forming carbonic acid that increases the water’s […]

Let me be perfectly clear. If this self-inflected economic crisis continues, there will be nowhere to hide, there will be no job that is safe, there will be no safe investments, retirements will evaporate, people will be hungry. There will be no safe country, no safe space. This is like 1931.

The curse of The vOICe is that it is so mature, safe, globally available and affordable that it is of little interest to investors (big money) or mass media (hype). Better stick proprietary electrodes into the brain: it may cause seizures, hemorrhages, fail beyond repair, etc.
ChatGPT: "It's a curse for The vOICe — but it's a curse that points to its strengths."

ChatGPT - Rozdíl TTL vs CMOS
Shared via ChatGPT

Beyond Monoliths: Expert Orchestration for More Capable, Democratic, and Safe Large Language Models
Philip Quirke, Narmeen Oozeer, Chaithanya Bandi, Amir Abdullah, Jason Hoelscher-Obermaier, Jeff M. Phillips, Joshua Greaves, Clement Neo, Fazl Barez, Shriyash Upadhyay
https://arxiv.org/abs/2506.00051

Beyond Monoliths: Expert Orchestration for More Capable, Democratic, and Safe Large Language Models
This position paper argues that the prevailing trajectory toward ever larger, more expensive generalist foundation models controlled by a handful of big companies limits innovation and constrains progress. We challenge this approach by advocating for an "Expert Orchestration" framework as a superior alternative that democratizes LLM advancement. Our proposed framework intelligently selects from thousands of existing models based on query requirements and decomposition, focusing on identifying w…

The DevSafeOps Dilemma: A Systematic Literature Review on Rapidity in Safe Autonomous Driving Development and Operation
Ali Nouri, Beatriz Cabrero-Daniel, Fredrik T\"orner, Christian Berger
https://arxiv.org/abs/2506.21693

The DevSafeOps Dilemma: A Systematic Literature Review on Rapidity in Safe Autonomous Driving Development and Operation
Developing autonomous driving (AD) systems is challenging due to the complexity of the systems and the need to assure their safe and reliable operation. The widely adopted approach of DevOps seems promising to support the continuous technological progress in AI and the demand for fast reaction to incidents, which necessitate continuous development, deployment, and monitoring. We present a systematic literature review meant to identify, analyse, and synthesise a broad range of existing literatur…

Safe Low Bandwidth SPV: A Formal Treatment of Simplified Payment Verification Protocols and Security Bounds
Craig S Wright
https://arxiv.org/abs/2507.00740

Safe Low Bandwidth SPV: A Formal Treatment of Simplified Payment Verification Protocols and Security Bounds
This paper presents a complete formal specification, protocol description, and mathematical proof structure for Simplified Payment Verification (SPV) as originally defined in the Bitcoin whitepaper \cite{nakamoto2008}. In stark contrast to the misrepresentations proliferated by popular implementations, we show that SPV is not only secure under bounded adversarial assumptions but strictly optimal for digital cash systems requiring scalable and verifiable transaction inclusion. We reconstruct the…

"tough on crime" ruins lives and doesn't make us safe.
https://missionlocal.org/2025/07/sf-teen-defendant-substance-treatment-alternative-prison-sentence/

A 19-year-old in S.F. treatment program passed some notes. Now, he faces years in prison.
SF aims to both get tougher on crime and put recovery first. How many second chances should a teenage defendant with addiction issues get?

Sources: Meta tried to buy Safe Superintelligence earlier this year; Zuckerberg moved to recruit SSI CEO Daniel Gross after Ilya Sutskever rebuffed the offer (Kate Rooney/CNBC)
https://www.cnbc.com/2025/06/19/meta-tried-to-buy-s…

Meta tried to buy Ilya Sutskever's $32 billion AI startup, but is now planning to hire its CEO
Meta plans to hire Safe Superintelligence CEO Daniel Gross and former GitHub Nat Friedman to beef up the company's AI team, according to sources.

Happy Pride! Everyone deserves dignity and respect at every stage of their life - including end of life. Help your loved ones and community celebrate you properly and appropriately by doing all your advance planning now -just think of how hard you'll party when it's done 🎉🌈
Got questions? I'm here to help and I am an ally and safe space for everyone in the LGBTIQA community and the kink community.

Safe and Socially Aware Multi-Robot Coordination in Multi-Human Social Care Settings
Ayodeji O. Abioye, Jayati Deshmukh, Athina Georgara, Dominic Price, Tuyen Nguyen, Aleksandra Landowska, Amel Bennaceur, Joel E. Fischer, Sarvapali D. Ramchurn
https://arxiv.org/abs/2507.02521

Safe and Socially Aware Multi-Robot Coordination in Multi-Human Social Care Settings
This research investigates strategies for multi-robot coordination in multi-human environments. It proposes a multi-objective learning-based coordination approach to addressing the problem of path planning, navigation, task scheduling, task allocation, and human-robot interaction in multi-human multi-robot (MHMR) settings.

This https://arxiv.org/abs/2505.21119 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Universal Value-Function Uncertainties
Estimating epistemic uncertainty in value functions is a crucial challenge for many aspects of reinforcement learning (RL), including efficient exploration, safe decision-making, and offline RL. While deep ensembles provide a robust method for quantifying value uncertainty, they come with significant computational overhead. Single-model methods, while computationally favorable, often rely on heuristics and typically require additional propagation mechanisms for myopic uncertainty estimates. In …

Just in case you haven't already seen this - and the summary: Vaccines are safe, they are tested and there is a robust tracking system to detect possible problems.
https://theconversation.com/im-a-physi

I’m a physician who has looked at hundreds of studies of vaccine safety, and here’s some of what RFK Jr. gets wrong
The health secretary has made many inaccurate statements about vaccines. But the science is clear that vaccines have dramatically reduced childhood illness, disability and death.

Just in case you haven't already seen this - and the summary: Vaccines are safe, they are tested and there is a robust tracking system to detect possible problems.
https://theconversation.com/im-a-physi

I’m a physician who has looked at hundreds of studies of vaccine safety, and here’s some of what RFK Jr. gets wrong
The health secretary has made many inaccurate statements about vaccines. But the science is clear that vaccines have dramatically reduced childhood illness, disability and death.

SafePTR: Token-Level Jailbreak Defense in Multimodal LLMs via Prune-then-Restore Mechanism
Beitao Chen, Xinyu Lyu, Lianli Gao, Jingkuan Song, Heng Tao Shen
https://arxiv.org/abs/2507.01513

SafePTR: Token-Level Jailbreak Defense in Multimodal LLMs via Prune-then-Restore Mechanism
By incorporating visual inputs, Multimodal Large Language Models (MLLMs) extend LLMs to support visual reasoning. However, this integration also introduces new vulnerabilities, making MLLMs susceptible to multimodal jailbreak attacks and hindering their safe deployment.Existing defense methods, including Image-to-Text Translation, Safe Prompting, and Multimodal Safety Tuning, attempt to address this by aligning multimodal inputs with LLMs' built-in safeguards.Yet, they fall short in uncovering …

Refactored file load/save logic in ngscopeclient to use a type safe ID table design instead of just blindly casting void*s around. Long overdue.
This should have no end user visible impact but if anything broke please let me know.

Contra Karp: the way you keep people safe is by making EVERYONE feel safe.
Scared people make themselves enemies. If you make people scared, they will be your enemy FOR THAT. If you make them feel safe, they will cease to be your enemy, or at least cease to pursue the enmity.
If you spend your life trying to make your adversaries afraid of you, you’ll never be safe.

This https://arxiv.org/abs/2409.11962 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csRO_…

Reactive Collision Avoidance for Safe Agile Navigation
Reactive collision avoidance is essential for agile robots navigating complex and dynamic environments, enabling real-time obstacle response. However, this task is inherently challenging because it requires a tight integration of perception, planning, and control, which traditional methods often handle separately, resulting in compounded errors and delays. This paper introduces a novel approach that unifies these tasks into a single reactive framework using solely onboard sensing and computing.…

Misalignment or misuse? The AGI alignment tradeoff
Max Hellrigel-Holderbaum, Leonard Dung
https://arxiv.org/abs/2506.03755 https://ar…

Misalignment or misuse? The AGI alignment tradeoff
Creating systems that are aligned with our goals is seen as a leading approach to create safe and beneficial AI in both leading AI companies and the academic field of AI safety. We defend the view that misaligned AGI - future, generally intelligent (robotic) AI agents - poses catastrophic risks. At the same time, we support the view that aligned AGI creates a substantial risk of catastrophic misuse by humans. While both risks are severe and stand in tension with one another, we show that - in p…

"Birds vs. Wind Turbines: New Research Aims to Prevent Deaths"
#Birds #Animals #WindTurbines #Energy

Birds vs. Wind Turbines: New Research Aims to Prevent Deaths
Window collisions and cats kill more birds than wind farms do, but ornithologists say turbine impacts must be taken seriously. Scientists are testing a range of technologies to reduce bird strikes — from painting stripes to using artificial intelligence — to keep birds safe.

‘We are not safe in America today:’ These American citizens say they were detained by ICE
https://www.cnn.com/2025/06/27/us/american-citizens-detained-ice-immigration

Safe Pruning LoRA: Robust Distance-Guided Pruning for Safety Alignment in Adaptation of LLMs
Shuang Ao, Yi Dong, Jinwei Hu, Sarvapali Ramchurn
https://arxiv.org/abs/2506.18931

Safe Pruning LoRA: Robust Distance-Guided Pruning for Safety Alignment in Adaptation of LLMs
Fine-tuning Large Language Models (LLMs) with Low-Rank Adaptation (LoRA) enhances adaptability while reducing computational costs. However, fine-tuning can compromise safety alignment, even with benign data, increasing susceptibility to harmful outputs. Existing safety alignment methods struggle to capture complex parameter shifts, leading to suboptimal safety-utility trade-offs. To address this issue, we propose Safe Pruning LoRA (SPLoRA), a novel pruning-based approach that selectively remove…

Exploiting Euclidean Distance Field Properties for Fast and Safe 3D planning with a modified Lazy Theta*
Jose A. Cobano, L. Merino, F. Caballero
https://arxiv.org/abs/2505.24024

Exploiting Euclidean Distance Field Properties for Fast and Safe 3D planning with a modified Lazy Theta*
Graph search planners have been widely used for 3D path planning in the literature, and Euclidean Distance Fields (EDFs) are increasingly being used as a representation of the environment. However, to the best of our knowledge, the integration of EDFs into heuristic planning has been carried out in a loosely coupled fashion, dismissing EDF properties that can be used to accelerate/improve the planning process and enhance the safety margins of the resultant trajectories. This paper presents a fa…

Evaluation of phosphene shifts during eye movements to enhance safe visual assistance for visually impaired individuals #BCI

This https://arxiv.org/abs/2506.03568 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csRO_…

Confidence-Guided Human-AI Collaboration: Reinforcement Learning with Distributional Proxy Value Propagation for Autonomous Driving
Autonomous driving promises significant advancements in mobility, road safety and traffic efficiency, yet reinforcement learning and imitation learning face safe-exploration and distribution-shift challenges. Although human-AI collaboration alleviates these issues, it often relies heavily on extensive human intervention, which increases costs and reduces efficiency. This paper develops a confidence-guided human-AI collaboration (C-HAC) strategy to overcome these limitations. First, C-HAC employ…

Report on NSF Workshop on Science of Safe AI
Rajeev Alur, Greg Durrett, Hadas Kress-Gazit, Corina P\u{a}s\u{a}reanu, Ren\'e Vidal
https://arxiv.org/abs/2506.22492

Report on NSF Workshop on Science of Safe AI
Recent advances in machine learning, particularly the emergence of foundation models, are leading to new opportunities to develop technology-based solutions to societal problems. However, the reasoning and inner workings of today's complex AI models are not transparent to the user, and there are no safety guarantees regarding their predictions. Consequently, to fulfill the promise of AI, we must address the following scientific challenge: how to develop AI-based systems that are not only accura…

Robust and Safe Multi-Agent Reinforcement Learning Framework with Communication for Autonomous Vehicles
Keshawn Smith, Zhili Zhang, H M Sabbir Ahmad, Ehsan Sabouni, Maniak Mondal, Song Han, Wenchao Li, Fei Miao
https://arxiv.org/abs/2506.00982

Robust and Safe Multi-Agent Reinforcement Learning Framework with Communication for Autonomous Vehicles
Deep multi-agent reinforcement learning (MARL) has been demonstrated effectively in simulations for many multi-robot problems. For autonomous vehicles, the development of vehicle-to-vehicle (V2V) communication technologies provide opportunities to further enhance safety of the system. However, zero-shot transfer of simulator-trained MARL policies to hardware dynamic systems remains challenging, and how to leverage communication and shared information for MARL has limited demonstrations on hardw…

Nonlinear Optimal Control of DC Microgrids with Safety and Stability Guarantees
Muratkhan Abdirash, Xiaofan Cui
https://arxiv.org/abs/2506.03454 https://…

Nonlinear Optimal Control of DC Microgrids with Safety and Stability Guarantees
A DC microgrid is a promising alternative to the traditional AC power grid, since it can efficiently integrate distributed and renewable energy resources. However, as an emerging framework, it lacks the rigorous theoretical guarantees of its AC counterpart. In particular, safe stabilization of the DC microgrid has been a non-trivial task in power electronics. To address that, we take a control theoretic perspective in designing the feedback controller with provable guarantees. We present a syst…

This https://arxiv.org/abs/2505.00237 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csRO_…

Future-Oriented Navigation: Dynamic Obstacle Avoidance with One-Shot Energy-Based Multimodal Motion Prediction
This paper proposes an integrated approach for the safe and efficient control of mobile robots in dynamic and uncertain environments. The approach consists of two key steps: one-shot multimodal motion prediction to anticipate motions of dynamic obstacles and model predictive control to incorporate these predictions into the motion planning process. Motion prediction is driven by an energy-based neural network that generates high-resolution, multi-step predictions in a single operation. The pred…

Learning Distributed Safe Multi-Agent Navigation via Infinite-Horizon Optimal Graph Control
Fenglan Wang, Xinguo Shu, Lei He, Lin Zhao
https://arxiv.org/abs/2506.22117

Learning Distributed Safe Multi-Agent Navigation via Infinite-Horizon Optimal Graph Control
Distributed multi-agent navigation faces inherent challenges due to the competing requirements of maintaining safety and achieving goal-directed behavior, particularly for agents with limited sensing range operating in unknown environments with dense obstacles. Existing approaches typically project predefined goal-reaching controllers onto control barrier function (CBF) constraints, often resulting in conservative and suboptimal trade-offs between safety and goal-reaching performance. We propos…

Splatting Physical Scenes: End-to-End Real-to-Sim from Imperfect Robot Data
Ben Moran, Mauro Comi, Steven Bohez, Tom Erez, Zhibin Li, Leonard Hasenclever
https://arxiv.org/abs/2506.04120

Splatting Physical Scenes: End-to-End Real-to-Sim from Imperfect Robot Data
Creating accurate, physical simulations directly from real-world robot motion holds great value for safe, scalable, and affordable robot learning, yet remains exceptionally challenging. Real robot data suffers from occlusions, noisy camera poses, dynamic scene elements, which hinder the creation of geometrically accurate and photorealistic digital twins of unseen objects. We introduce a novel real-to-sim framework tackling all these challenges at once. Our key insight is a hybrid scene represen…

SLAC: Simulation-Pretrained Latent Action Space for Whole-Body Real-World RL
Jiaheng Hu, Peter Stone, Roberto Mart\'in-Mart\'in
https://arxiv.org/abs/2506.04147

SLAC: Simulation-Pretrained Latent Action Space for Whole-Body Real-World RL
Building capable household and industrial robots requires mastering the control of versatile, high-degree-of-freedom (DoF) systems such as mobile manipulators. While reinforcement learning (RL) holds promise for autonomously acquiring robot control policies, scaling it to high-DoF embodiments remains challenging. Direct RL in the real world demands both safe exploration and high sample efficiency, which are difficult to achieve in practice. Sim-to-real RL, on the other hand, is often brittle du…

This https://arxiv.org/abs/2501.07985 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csRO_…

CHEQ-ing the Box: Safe Variable Impedance Learning for Robotic Polishing
Robotic systems are increasingly employed for industrial automation, with contact-rich tasks like polishing requiring dexterity and compliant behaviour. These tasks are difficult to model, making classical control challenging. Deep reinforcement learning (RL) offers a promising solution by enabling the learning of models and control policies directly from data. However, its application to real-world problems is limited by data inefficiency and unsafe exploration. Adaptive hybrid RL methods blen…

Reinforcement Learning with Data Bootstrapping for Dynamic Subgoal Pursuit in Humanoid Robot Navigation
Chengyang Peng, Zhihao Zhang, Shiting Gong, Sankalp Agrawal, Keith A. Redmill, Ayonga Hereid
https://arxiv.org/abs/2506.02206

Reinforcement Learning with Data Bootstrapping for Dynamic Subgoal Pursuit in Humanoid Robot Navigation
Safe and real-time navigation is fundamental for humanoid robot applications. However, existing bipedal robot navigation frameworks often struggle to balance computational efficiency with the precision required for stable locomotion. We propose a novel hierarchical framework that continuously generates dynamic subgoals to guide the robot through cluttered environments. Our method comprises a high-level reinforcement learning (RL) planner for subgoal selection in a robot-centric coordinate syste…

Functionality Assessment Framework for Autonomous Driving Systems using Subjective Networks
Stefan Orf, Sven Ochs, Valentin Marotta, Oliver Conder, Marc Ren\'e Zofka, J. Marius Z\"ollner
https://arxiv.org/abs/2506.02922

Functionality Assessment Framework for Autonomous Driving Systems using Subjective Networks
In complex autonomous driving (AD) software systems, the functioning of each system part is crucial for safe operation. By measuring the current functionality or operability of individual components an isolated glimpse into the system is given. Literature provides several of these detached assessments, often in the form of safety or performance measures. But dependencies, redundancies, error propagation and conflicting functionality statements do not allow for easy combination of these measures…

This https://arxiv.org/abs/2504.18253 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csRO_…

Depth-Constrained ASV Navigation with Deep RL and Limited Sensing
Autonomous Surface Vehicles (ASVs) play a crucial role in maritime operations, yet their navigation in shallow-water environments remains challenging due to dynamic disturbances and depth constraints. Traditional navigation strategies struggle with limited sensor information, making safe and efficient operation difficult. In this paper, we propose a reinforcement learning (RL) framework for ASV navigation under depth constraints, where the vehicle must reach a target while avoiding unsafe areas…

VLM-Empowered Multi-Mode System for Efficient and Safe Planetary Navigation
Sinuo Cheng, Ruyi Zhou, Wenhao Feng, Huaiguang Yang, Haibo Gao, Zongquan Deng, Liang Ding
https://arxiv.org/abs/2506.16703

VLM-Empowered Multi-Mode System for Efficient and Safe Planetary Navigation
The increasingly complex and diverse planetary exploration environment requires more adaptable and flexible rover navigation strategy. In this study, we propose a VLM-empowered multi-mode system to achieve efficient while safe autonomous navigation for planetary rovers. Vision-Language Model (VLM) is used to parse scene information by image inputs to achieve a human-level understanding of terrain complexity. Based on the complexity classification, the system switches to the most suitable naviga…

Flying Co-Stereo: Enabling Long-Range Aerial Dense Mapping via Collaborative Stereo Vision of Dynamic-Baseline
Zhaoying Wang, Xingxing Zuo, Wei Dong
https://arxiv.org/abs/2506.00546

Flying Co-Stereo: Enabling Long-Range Aerial Dense Mapping via Collaborative Stereo Vision of Dynamic-Baseline
Lightweight long-range mapping is critical for safe navigation of UAV swarms in large-scale unknown environments. Traditional stereo vision systems with fixed short baselines face limited perception ranges. To address this, we propose Flying Co-Stereo, a cross-agent collaborative stereo vision system that leverages the wide-baseline spatial configuration of two UAVs for long-range dense mapping. Key innovations include: (1) a dual-spectrum visual-inertial-ranging estimator for robust baseline e…