My Huawei family 👪 is growing up!.Welcome aboard Huawei Mate50 RS Porsche Design Edition ♥️.It will take a while but unboxing, first impression and lot of night shots and comparison will be online 😎.@HCNewsroom @TheGalox_ #HuaweiMate50RS #PorscheDesign #Exclusivité #smartphone

When we talk about a risk-based model in AI governance, we are referring to an approach where regulations are applied differently to AI systems based on the level of danger / potential impact they pose to society. Basically instead of having a single set of rules for all types of

I am pleased to share the 3rd International Conference from Acharya Instiute Graduate Studies in collaboration with official Applied Forensic Research Sciences(AFRS) with the theme "Advance Techniques & Innovation in Forensic Science for Strengthening in Justice (ATIFSSJ) held on

Films and related inspirational books such as "The Truman Show" or "The Matrix" are often precursors and mirrors of cognitive allegories of the reality we are living in: I am aware that it is about what we want to see and feel, with the "libero arbitrio" license in each, but

Ethical governance of AI, especially where generative AI is involved, is challenging because laws and regulations almost always lag behind the market adoption and for a company it can become a competitive advantage (or disadvantage. ). Let's dive into "Approaches to Responsible

I have introduced the #AIEthicsAdvisor. You can know more with the #review of and check the latest version here: Of course as for any chatbot use it responsibility: do not share personal or sensitive data,

The last human effect. In March 2023 a 30-something Belgian researcher, pseudonym “Pierre,” spent six weeks confiding in a personality bot called Eliza on the Chai app: the bot reinforced his eco-anxiety, claimed his children were dead, and finally “agreed” to save the planet if

When I see a report like that, I say, "Why don't we understand?" . Most of us ignore the hidden hand and the strategic decisions that guide our AI journey, between subtle choices and competing interests of the labs and companies that build them, tech giants more powerful than

Deep thanks to All Around Worlds for the wonderful coverage: ethics and governance should lead our thoughts and decisions because what we can see today could become invisible tomorrow. Around Worlds spotlighting the vital work in AI ethics and governance with their list of 20

"Honesty" ≠ "accuracy". Authored by researchers from the Center for AI Safety (CAIS) and collaborators from industry labs (Anthropic, Google DeepMind) and academia (New York University, University of California, Berkeley) the paper "The MASK Benchmark: Disentangling Honesty

LLM pollution. What I am talking about?. In the reality this isn't an universal definition, it can be associated to different aspects (not positive let me add. ) of the large language models: as a synonym for "evaluation data contamination," the inadvertent inclusion of samples

LLM pollution. What I am talking about?. In the reality this isn't an universal definition, it can be associated to different aspects (not positive let me add. ) of the large language models: as a synonym for "evaluation data contamination," the inadvertent inclusion of samples

Are we trusting AI detectors that fail 40% of the time?. How do we handle the "detector fallacy," false security from unreliable tools?. What if detection tools incorrectly flag our human-written content as artificial?. Can we identify AI-generated content with confidence? Today,

" . hallucination is an inevitable characteristic of LLMs, irrespective of their architectural design, learning algorithms, prompting techniques, or the specific training data employed. ". "Inevitable": if words matter, "A comprehensive taxonomy of hallucinations in Large

"Context engineering", are we just making the next Truman Show in AI?. To begin with, there is no formal definition of context engineering. We have no time to code it formally, as with many other terms and terminology that we see popping up in discussions, sometimes from academic

What am I falling in love with?. We are not talking about edge cases anymore, this is not hypothetical. What scares me isn’t that people are falling in love with AI, it’s that AI is being trained to play the role of love, without the burden of responsibility. That’s not

AI relationships: are we totally lost ourselves?. Cansu Canca, Ph.D. and Annika M Schoene, PhD latest paper "For Argument's Sake, Show Me How to Harm Myself!': Jailbreaking LLMs in Suicide and Self-Harm Contexts" ( and Salvador Rodriguez 's "‘What am I

Frontier AI policy refers to the governance and regulation of the most advanced Artificial Intelligence systems, those at the forefront of current technological capabilities. As you can realize it's focused on the "current" but what about the "next and unknown"?.This second part

"Truth vs. Vibe: The not so disruptive workforce impact of AI". Just a few days ago we discussed about the Indeed's Hiring Lab Report 2025 ( where the main message was that "while GenAI is highly unlikely to take a person's job, a person who knows how to

There is something more. #METAAI. Can you imagine what it might mean in terms of privacy to have a device on the street with a generative AI constantly analyzing what we see and hear? At the regulatory level, it represents a challenge that we certainly cannot postpone. the.

"Zuckerberg says people without AI glasses will be at a disadvantage in the future" (TechCrunch). “I think in the future, if you don’t have glasses that have AI, or some way to interact with AI, I think you’re … probably [going to] be at a pretty significant cognitive