BREAKING: THE STARGATE JOINT VENTURE BETWEEN OPENAI, ORACLE AND SOFTBANK HASN’T STAFFED UP AND ISN’T BUILDING OPENAI’S DATA CENTERS, PER THE INFORMATION ...
France Bank-Account Database Breach: 1.2 Million Records Stolen — An attacker breached a French database that catalogs bank-account details and exfiltrated roughly 1.2 million records, according to security reports. Authorities are investigating the scope and potential fraud risks as…
📰 Claude C Compiler Reveals AI’s Potential and Pitfalls in Software Engineering
Anthropic’s Claude AI, using parallel models on Opus 4.6, has built a functional C compiler — a milestone that showcases AI’s growing ability to replicate complex software systems. Yet expert...
I just listened to two guys who talk faster than Ben Shapiro. Dario Amodei from anthropic and Dwarkish Patel. It's very urgent we spend trillions of dollars on compute, to be the first to have a 'data center full of geniuses'.
Time to lay off the coke, let your feet touch the ground again.
he shared his findings with The Verge, which quickly contacted DJI to report the flaw.
the issue has been “resolved,”
cybersecurity experts who have long-warned that internet-connected robots and other smart home devices present attractive targets for hackers.
In addition to calling Bixby or Gemini, Galaxy S26 users will be able to call Perplexity by saying, “Hey, Plex.” The integration of Perplexity into Galaxy AI is just one element of the company's embrace of a "multi-agent ecosystem." Often, people will use…
Great, anthropic. Yes indeed uses SHGetKnownFolderPath, that's true. Maybe some things are fine to not vibe code some of the time... github.com/anthropics/c...
PayPal reports a nearly six-month data exposure in its Working Capital loan app. Sensitive customer data was compromised, including social security numbers, and even unauthorized charges alternativeto.net/news/2026/2...
In recent interviews, Sam Altman said AI's adoption faces more resistance than he expected, while Jensen Huang warned the "doomer narrative" may be winning (David Streitfeld/New York Times)
What would be a convincing use case for agentic AI in the context of supply-chain management? Samir Saci outlines a potential approach in great detail.
Here is a good example of technofascism, & how the rich use propaganda "news" to control your opinion about a topic.
All this AI does is give you honest responses. The other ones frame every reply from a rich psychopath perspective. They (ChatGPT) also lie. www.esecurityplanet.com/artificial-i...
AI is about to expose weak BI architecture. "DirectQuery" collapses under machine curiosity. Decision-aligned design is the only way forward. #artificialintelligence
Samsung Galaxy S26 Release Could Finally Challenge Pixel Loyalty With Radical Hardware Upgrades
For nearly a decade, Google’s Pixel lineup has built a quiet but powerful cult following. Clean Android, intelligent AI features, and dependable cameras have kept loyal users firmly rooted inside…
Google Blocked 1.75 Million Malicious Apps from Entering into the Play Store AI-powered security systems blocked over 1.75 million malicious or policy-violating apps from reaching the Play Store in...
Sources: Dell, Lenovo, and other PC makers are working with Nvidia on laptops using the Arm-based Nvidia-MediaTek system-on-a-chip, which could come in H1 2026 (Yang Jie/Wall Street Journal)
Should you use Doctrine DQL or Native SQL in Symfony? Learn when each wins with benchmark results, DTO patterns, and real production scenarios. #symfony
The FBI has named a series of Wi-Fi routers as being unsafe. If you're using a router from this brand, you might want to check if yours is one of the models.
i dislike gemini 3.1 pro
it still can't do tool call correctly.
the rate limit in anigravity is so much lower with that model.
and it is still not half as good as claude in actual use.
gemini 3.1 is designed for their crappy web chat app, not actual agentic system
PaCMan - Partition-Code Masking for Combined Security (Fabian Buschkowski, Jakob Feldtkeller, Tim Güneysu, Elisabeth Krahmer, Jan Richter-Brockmann, Pascal Sasdrich) ia.cr/2026/343
Anthropic Debuts Security Tool as Cybersecurity Stocks Fall Cybersecurity software stocks fell last week after Anthropic unveiled a new security feature for its Claude model. That’s according to ...
"Convenience must exist alongside privacy, not instead of it." Why @EFF.org is targeting major platforms to stop stalling and implement end-to-end encryption by default
A Coding Guide to Instrumenting, Tracing, and Evaluating LLM Applications Using TruLens and OpenAI Models In this tutorial, we focus on building a transparent and measurable evaluation pipeline for...
The Dark Reset: Survival Before the Silence is a practical preparedness guide designed to help families stay safe, organized, and self-reliant during unexpected disruptions. Built on real-world logistics experience and research visit us- en-us-en-darkreset.com
Looking for a very very affordable video editor to help w some overflow work on my YouTube channel. Need someone with STRONG editorial judgement and the ability to edit news footage on tech policy and complicated topics with an eye for detail. Student or teen is fine, but they need to be reliable!
Experian’s new fraud forecast warns agentic AI, deepfake job candidates and cyber break-ins are top threats for 2026.. www.experianplc.com/newsroom/pre...
An AI agent created by an OpenAI employee accidentally sent over $441,000 worth of cryptocurrency to a stranger who had requested just $310. The agent, named Lobstar Wilde, was tasked with growing a $50,000 investment but lost its entire holdings in the errant transaction. The recipient promptly…
Samsung will add Perplexity to Galaxy AI on the upcoming S26 series; users can launch the Perplexity agent by saying "Hey Plex" or with a physical helper button (Cheyenne MacDonald/Engadget)
The FBI has named a series of Wi-Fi routers as being unsafe. If you're using a router from this brand, you might want to check if yours is one of the models.
v1.0.1 of flutter_story_encoder Production-grade, hardware-accelerated video export engine for social media stories on iOS (AVFoundation) and Android (MediaCodec).
Apply Now: $360,000 for AI4Good AgriTech Farmer Solutions
Traditional agriculture extension workers are overwhelmed. Advisor-to-farmer ratios exceed 1:1,000 across developing countries. Government extension systems are unable to meet the needs of hundreds of millions of farmers. Digital solutions…
Four AIs now argue before replying. xAI's Grok 4.2 beta uses a multi-agent setup and claims a 65% drop in hallucinations. Weekly public updates beat Claude and Gemini. Strong at reasoning and coding. Big promise, real concerns. Will you test it?
Keep your East Thompson home safe with Lynx Systems LLC! 🏠 From smart alarms and security cameras to fire protection and home automation, our home security systems provide reliable, easy-to-use protection and peace of mind.
Bluesky is ridiculously well-positioned for the agent internet. Everything signed = agents can build up un-fakeable reputations. newsletter.squishy.computer/p/llms-break...
The hidden security cost of treating labs like data centers In this Help Net Security interview, Rich Kellen, VP, CISO at IFF, explains why security teams should not treat OT labs like IT environme...
Exclusive: Epstein hid secret files in storage units across US
He leased at least one unit from 2003, when he was part of a Florida social set that included Donald Trump. Credit card receipts obtained by The Telegraph show regular storage payments continued until 2019
In addition to calling bixby or Gemini, Galaxy S26 users will be able to call Perplexity by saying "hello plex.” Perplexity's integration into Galaxy AI is just one element of the company's adoption of a "multi-agent ecosystem." Often, people will use…
Balancing Privacy with Innovation is key. It's just not business model transformation but starts with a mindset transformation that we value our own data.
作者因長期關注隱私，動手做了名為 Bluehood 的 Bluetooth 掃描器，想弄清楚「只要把 Bluetooth 開著」到底會外洩多少生活線索。契機之一是比利時 KU Leuven（魯汶天主教大學）研究團隊近期揭露 WhisperPair（CVE，Common Vulnerabilities and […]
Alpha Data highlights the role of cybersecurity in an exclusive emt interview Alpha Data highlights the strategic role of cybersecurity as a business enabler in an exclusive interview with emt. The...
Why the Samsung Galaxy S26 Ultra is the Ultimate Creator’s Tool The Samsung Galaxy S26 Ultra represents a pivotal step in smartphone evolution, blending AI-powered features, enhanced audio capabi...
@johnmu.com
Can you see the highlighted part in the AI overview. Is there any bug or google is trying something new with an AI overview. Can you please explain.
"SuperClaw addresses a growing blind spot in enterprise AI adoption: agents are routinely deployed with broad tool access and high privileges, yet most organizations skip structured security validation entirely before going live."
Google's Gemini 3.1 Pro just dropped. Scored 77.1% on ARC-AGI-2 — more than 2x its predecessor. Beats GPT-5.2 and Claude Opus 4.6 on reasoning. You can adjust "thinking depth" in 3 levels. Free to try on Google AI Studio.
When an AI chatbot arbitrarily refuses to answer a prompt, is it a safety feature or outright censorship? The latest from @accessnow.org explores the hidden dangers of LLM refusal mechanisms and how "alignment" often lacks a human rights lens
OpenAI and Jony Ive plan to release a smart speaker in early 2027.
The hardware lacks a screen to prevent digital distraction.
An integrated camera uses facial recognition to identify users and their surroundings. macrumors.com
Stopping active attacks isn't about the event. It's about the full journey.
In general, most cloud vendors are just going to see what's in their own yard. Events firing off as normal. But when you look not at the yard but the whole neighborhood, you see the actual story. https://loom.ly/44b0_pk
🆕 KDAB Whitepaper: Windows CE to Embedded Linux Migration
This is a practical guide for migrating business logic, UI & OS/hardware interactions covering C++, C#, Qt, Linux, and team prep for a smooth transition. #Cpp#QtDev#Embedded#Linux#Windows#Wince
Good discussion between @mikell.bsky.social (worked with her in the early days of Rethink Robotics) and @bheater.bsky.social
of A3, about the reality of building and deploying robots. Mikell has been at many robotic startups and more recently Amazon Robotics, and GM. www.youtube.com/watch?v=KNLe...
Monday’s Den: Bitcoin miners sold every satoshi for AI GPUs. Spotify devs haven’t coded since December. CS enrolment cratering. The Pentagon called Anthropic a "supply chain risk" for wanting humans in the weapons loop. Normal Monday.
-AI hacking campaign breaches 600+ Fortinet devices
-Ivanti hacked via its own VPN in 2021
-China hacks Italy’s police force
-Wikipedia bans ArchiveToday for DDoS attacks
-Bug bounty offered to de-Amazon Ring cameras
One might connect such enthusiasm (ex Tory politicians George Osborne and Rishi Sunak pimping digital asbestos, AI ) to their newly minted jobs with OpenAI and Anthropic, respectively, but we are bidden to banish such cynicism from our minds.
A look at the rising popularity of "WorkTok" videos, where employees film their daily work routines on TikTok and Instagram, particularly among Gen Z viewers (Kimi Chaddah/Financial Times)
It has only been two months since Manage My Health was targeted in a data breach and now another online health platform has had its records "incorrectly modified'. Latest from me
#Meta and Other Tech Firms Put Restrictions on Use of #OpenClaw Over #Security Fears - www.wired.com/story/opencl... "Security experts have urged people to be cautious with the viral agentic AI tool, known for being highly capable but also wildly unpredictable." #ai#agents
📰 Open-Source Framework Enables Local Models to Match Gemini 3.1 Pro Performance
A new open-source scaffolding framework is enabling locally deployed AI models to achieve reasoning capabilities rivaling Google’s Gemini 3.1 Pro, challenging the dominance of proprietary cl...
Galaxy S22 owners are dealing with endless bootloops after the latest patch Samsung's February 2026 update is bricking Galaxy S22 phones for some users. Credit: C. Scott Brown / Android Authori...
New Technical Briefing: Digital Fingerprinting - blog.citp.princeton.edu/2026/02/20/n... "The brief walks through how fingerprinting works technically, who provides the technology, and why it is spreading."
AI Automation in Cybersecurity: Revolutionizing Defense 2026
AI automation is revolutionizing cybersecurity. See how AI empowers 2026 defenses, fighting advanced threats with predictive, autonomous security systems.AI automation is revolutionizing cybersecurity in 2026. Discover how AI powers both…
There are many extremely important questions Gen AI providers must ask about data privacy and consent. I certainly do not consent to being recorded, unless specifically asked, and you might want to tell your work if you feel similarly www.ft.com/content/b24e...
Resistance Training Prescriptions Generated by ChatGPT and Google Gemini: Cross-Sectional Comparative #Study
Abstract (preprint) #openscience#PeerReviewMe#PlanP
Should we wait for an AI Chernobyl to start taking security seriously? A brilliant conversation on why moving deliberately & securing Confidentiality, Integrity and Availability is far more important than unchecked acceleration, via @alexmayoralb.bsky.social
Sam Altman claims ChatGPT's adult mode will 'be able to safely relax the restrictions' of the chatbot, but firing a critic of the plan is a reason to be wary www.techradar.com/ai-platforms...
First‑ever Mobile Hacking Conference, happening next week — fully online and completely free.
I’ve summarized everything you need to know in one place — speakers, talks, CTFs, giveaways, free courses, and more. www.mobile-hacker.com/2026/02/23/t...
How Buddy Rizer, an official in Virginia's Loudoun County, helped build the world's largest data center market, with ~200 buildings spanning ~49M square feet (David Uberti/Wall Street Journal)
AWS describes the hacker as an “unsophisticated” individual or small group armed with AI tools that helps them achieve operational scale to commit crime, something that would have previously required a significantly larger and more skilled team.
Ukrainian hackers infiltrated Russian drone operator systems for six months, exposing how Russia weaponized Belarusian civilian infrastructure to strike Ukraine and test attack routes into NATO countries euromaidanpress.com/2026/02/22/r...
User accidentally gains control of over 6,700 robot vacuums while tinkering with their own device to enable control with a PlayStation controller — security flaw reveals floor plans and live vide...
AI is transforming insider risk as data moves across agents, apps, and collaboration tools—forcing CISOs to rethink governance fast. #aiinsiderriskmanagement
They did it again. After the official Raspberry Pi SD card and SSD, there are now official USB flash drives. Is that worth more than a generic USB stick with the Raspberry Pi logo on it? Let’s find out. Raspberry Pi is now selling official USB... #raspberrypi #linux
Protect your data with Immutable Snapshots for NAS secure, recoverable, and tamper-proof against ransomware or admin errors. Ensure business continuity and compliance.
Japanese tech giant Advantest suffered a ransomware attack, leading to data encryption. They confirmed no client data was affected. This incident highlights ongoing cybersecurity challenges for tech firms. #CyberSecurityAlert
This piece is fascinating. Amazon is encouraging the use of AI to draft its famous "six-pagers" - which undermines the whole purpose of writing the document in the first place. Writing is thinking; pressing an AI button is not. www.bigtechnology.com/p/writing-cr...
What Becomes Of The Global Creator Economy Amid Growing AI Slop?
In December 2024 when OpenAI released its text-to-video model Sora, it signalled that the creator economy had entered its synthetic age. Around the same time, tools like Runway, Pika, and image generators such as DALL·E and…
Aren’t we lucky that universities haven’t gotten carried away with the hype and taken a careful approach to AI?
Imagine if they had decided to push AI into our teaching and work life!
All praise to our leaders in higher ed for not falling for hype, hallelujah and praise the goddess
Una base de dades vinculada a la verificació d’identitat mal configurada exposa les dades de 1.000 milions de persones. Noms, adreça física, números d’identificació, telèfons, correus electrònics...
Verificar l'edat segons com és més insegur que no fer-ho. www.tomsguide.com/computing/on...
If you do not use Cyber Tech on time and Legally I can almost Guarantee that the E.U. will lose an equivalent of more than 1 Billion Euros as the Chinese and others Develop so much faster.
I have you know that Neglecting and Abandoning Advanced Science always has Consequences.
Police in Poland have dismantled an organized group that used phishing to steal over 100,000 Facebook credentials, which were then used for further fraud.
Join us ahead of International Women’s Day for a webinar which will spotlight women in cyber security and explore ways to engage more women in this critical field.
Speakers:
Dr Mary Haigh, Director of Digital Delivery, BAE Systems plc
Prerana Joshi, Research Fellow at RUSI
As Anthropic releases its most autonomous agents yet, a mounting clash with the military reveals the impossible choice between global scaling and a “safety first” ethos
Economists at Morgan Stanley and JPMorgan report the AI boom contributed "basically zero" to US economic growth in 2025, challenging claims of up to 92% growth (Shira Ovide/Washington Post)
Carefull, it's begun, the british Jews will turn against the greens like they turned again Corbyns labour party.
If you want to be elected don't support Palestinians
Manchester a Jewish city
zack you know this.you lived there.
ohhhh man, this piece is just straight up panic bait garbaaaaage. literally every. single. example. of supply chain disruption they identify & describe *has nothing to do with AI*.
Figma Make lets users choose between Claude 4.6 and Gemini 3 for AI-driven prompt-to-code UI generation, producing interactive prototypes with React and Tailwind code.
IS-995: ICS Position Foundation (2025) provides emergency responders with a practical introduction to ICS duties, responsibilities, and organizational capabilities.
EMI training: https://ow.ly/Fj7K50YjLh8
Learn more about IMTA at: https://www.ahimta.org/membership
Allan “Ransomware Sommelier” Liska@ransomwaresommelier.com
On the one hand this is bad. On the other other hand, this is bad but I’ve seen experienced IAB teams carry out similar campaigns in that time frame. The difference is EXPERIENCED and TEAMS.
This could serve as blueprint for anyone who wants to be an IAB.
Tom's Hardware - Article
"MIT-developed 3D printer can output a fully functional electric motor in a single process — team only needed to magnetize the linear motor after printing, motors cost just 50 cents each"...
1/ The dispute between Anthropic and DOD over the limits of using AI in warfare is escalating. I’ll have more to say on what these limits should be in a @brennancenter.org report out next month. But for now, here’s what you need to know, and what’s truly at stake 🧵:
Zero-Day Exploit Chain in Latest AI Supply Chain Attack: How to Detect and Mitigate + Video
Introduction: A sophisticated supply chain attack targeting AI/ML development pipelines has been uncovered, leveraging vulnerabilities in popular open-source libraries and CI/CD misconfigurations. This…
Can Agentic AI boost confidence in privileged access management
How Can We Enhance Security with Effective Non-Human Identity Management? Can your organization’s security posture handle the growing complexity of machine identities and their secrets? With the increasing deployment… #hackernews#news
The article doesn’t ask Anthropic what the DoD is requesting from them, nor does Wired speculate. I speculate it’s asking Anthropic to target Americans who write things the President doesn’t like on social media.
Benjamin Dowling, Prosanta Gope, Mehr U Nisa, Bhagya Wimalasiri
Drawing the LINE: Cryptographic Analysis and Security Improvements for the LINE E2EE Protocol https://arxiv.org/abs/2602.18370
Said Varlioglu, Nelly Elsayed, Murat Ozer, Zag ElSayed, John M. Emmert
Detecting PowerShell-based Fileless Cryptojacking Attacks Using Machine Learning https://arxiv.org/abs/2602.18285
Kevin Hermann, Sven Peldszus, Thorsten Berger
Many Tools, Few Exploitable Vulnerabilities: A Survey of 246 Static Code Analyzers for Security https://arxiv.org/abs/2602.18270
Cathrin Schachner, Jasmin Wachter
Can AI Lower the Barrier to Cybersecurity? A Human-Centered Mixed-Methods Study of Novice CTF Learning https://arxiv.org/abs/2602.18172
Interesting find and I wasn't even looking for it... Pathetic over-optimized review shows up in my feedly. Gobs of content. Too much. No author listed... Sets off my internal "this looks AI-generated content" flag. Check the page, yep, 99% probability of AI.
Deepfakes can cost real money: 'Is my boss really asking me to transfer $30k?' Journalists now use probability to help spot fakes. Their framework translates to business ... bryl.us/2mj8#Deepfakes#Cybersecurity
Phan The Duy, Nghi Hoang Khoa, Nguyen Tran Anh Quan, Luong Ha Tien, Ngo Duc Hoang Son, Van-Hau Pham
PenTiDef: Enhancing Privacy and Robustness in Decentralized Federated Intrusion Detection Systems against Poisoning Attacks https://arxiv.org/abs/2602.17973
Ivanti has released this report warning that cyberattacks are accelerating in number and sophistication, and organisational defences are struggling to keep pace. www.ivanti.com/resources/re...
Last year I put together a list of products everyone absolutely needed to own in 2025. It included basic stuff, AirTags, GaN chargers, and even some slightly complex gadgets like NAS devices to help you cut the cord on cloud storage subscriptions. This year's list expands on the same philo..
• A designer’s framework for better AI prompts
• AI fatigue is real and nobody talks about it
• From Claude Code to Figma
• Figma Console MCP
• New Make connectors
• $FIG 2025 Q4 earnings
• Why demand for designers is on the rise
Unilever and Google Cloud announced a five-year AI partnership focused on agentic commerce and an AI-first data backbone. Financial terms weren’t disclosed.
There's a conspiracy theory that suggests that since around 2016 most web activity is automated. This is called Dead Internet Theory, and while I think they may have jumped the gun by a few years, it's heading that way now that LLMs can simulate onli...
If you use Discord, your activity in both public messages and voice channels might have been scraped and sold to AI companies and law enforcement agencies for as little as $5:
stories like this will mention the AI "doomer narrative" (SkyNet will kill us all, often pushed by industry insiders), but not mention it pre-empts more realistic criticism (Grok's data center in Memphis is aiming its pollution directly at black people) the industry doesn't want to talk about
#TechRadar Anthropic claims half of its agent tool calls are to do with software engineering - so are developers letting AI take over? https://techrad.ar/hLnp#Pro
Like Windows XP & Windows 7 devices have a useful life. The FBI is warning older #Wi-Fi routers from the early 2000's that their devices are at risk of #cyberattacks. Like #patching updating devices is a vital part of #cyber. Turn to our global team with questions www.slashgear.com/2104404/wifi...
Google for Education, with ISTE and ASCD, will offer free Gemini-focused AI literacy training to all 6 million U.S. K–12 and higher-education faculty—covering 74 million students—using microcredentials, bite-sized modules, and tools like Gemini, NotebookLM, and Guided Learning.
Anthropic's code scanner sent cyber stocks crashing. Okta. SailPoint. Down 9%+. For a tool that doesn't touch identity security at all. Attackers have two doors — code vulnerabilities and identity abuse. AI just slammed one of them. The other is still wide open. 🔗
Anthropic launched Claude in PowerPoint as a Microsoft AppSource add-in (research preview) that embeds Claude into active PowerPoint to generate native editable slides, support three workflows, read slide masters, and warn of beta security and compliance limitations.
Attackers cloned Gemini with 100K+ prompts. Your GPU monitoring vendor sees every batch size and training pattern. For proprietary models: telemetry stays on-prem or your attack surface is someone else's infrastructure.
The BCI market in China is ahead of the market in the US, and is accelerating away. The brain computer interfaces that ultimately improve the lives of people in the US will be developed and made in China, and the intellectual property profits will also go to China. techcrunch.com/2026/02/22/c...
To ensure your AI project evaluation is robust, there are several components you need to plan for in advance. @stephaniekirmer.com outlines best practices and zooms in on the most crucial preparatory aspects of the process.
If you are watching for signs of AI recursive self-improvement, take note of OpenAI's latest claim on #FirstProof.
It's been messy, and the proofs are still not validated, but it points to genuine novel discovery, potentially fully automated.
AI अब मैलवेयर में भी इस्तेमाल हो रहा है।
PromptSpy Malware Android स्क्रीन डेटा चोरी कर Google Gemini से फोन कंट्रोल करता है।
ESET रिसर्च में खुलासा।
सुरक्षा के लिए Google Play Protect ऑन रखें और unknown apps से दूर रहें।
अधिक जानकारी के लिए 👇 www.tanatangyan.com/2026/02/prom...
OpenAI has launched initiatives to support AI research in India, collaborating with local institutions to enhance access to AI tools and education. This effort aims to foster innovation and empower the next generation of AI talent.
Last night and this morning I have been toying with the idea of the paid tiers and whether they're worth it. In the proces I told Gemini about my context and it went from advising me the pro tier at 170 CHF per year, to going for the "free" Google […]
Last night and this morning I have been toying with the idea of the paid tiers and whether they're worth it. In the proces I told Gemini about my context and it went from advising me the pro tier at 170 CHF per year, to going for the "free" Google AI…
Lorenzo Franceschi-Bicchierai@lorenzofb.bsky.social
NEW: There have been a seemingly endless series of critical flaws and cybersecurity incidents related to Ivanti's VPN appliances in the last few years.
Turns out there was a major one in 2021 that wasn't reported until now, according to Bloomberg.
Despite rapid progress in AI’s ability to build complex software and solve advanced physics problems, the ubiquitous format of PDF remains something of a grand challenge.
OpenAI has inked a deal with four of the world’s biggest consulting firms to help push more firms past the sticky AI pilot phase and into full‑scale, global deployment, stepping up its race with Anthropic to capture enterprise clients.
Stop calling AI prompt engineering a business – together with @savoirr.bsky.social's senior advisor and angel investor Jean-Marc Hensch @sosicles.bsky.social, we discussed the impact of Gen AI on SaaS business models, especially after the so called #SaaSpocalypse early February 2026.
Brussels’ privacy reforms stumble out the gate Countries shoot down a core part of Brussels' privacy reform at the earliest opportunity. BRUSSELS — The European Union’s sweeping privacy r...
‘Over the past year, the running joke in the data center industry—even inside OpenAI—has been that the $500 billion Stargate initiative was less of a master plan than a “vibe.”’ www.theinformation.com/newsletters/...
OpenAI announced on Friday that the company considered reporting the suspect in the Tumbler Ridge shooting — Jesse Van Rootselaar — to the Canadian police over their use of ChatGPT, but ultimately elected not to.
Open WebUI v0.8.4 just dropped with a HUGE performance boost. Anthropic API proxy & full model support. Critical RAG & security fixes. No breaking changes, just a better self-hosted AI experience!
Backed by Anthropic, a Super PAC Begins an Ad Blitz in Support of A.I. Regulation www.nytimes.com/2026/02/23/t...
AI's dangerous.On"The Pitt"a doctor introduced AI to record patients'records.Smart young doctor caught a dangerous error-2 conditions or meds SOUNDING alike,one wd be deadly,other not.
“The deteriorating threat landscape is driving a surge in private capital toward 'secure-by-design' systems and collaborative frameworks like the Colorado-based Space ISAC.” bisi.org.uk/reports/cybe...
Accounting firms leverage AI to expand online marketing Accounting firms can't simply focus on showing up in online search results; they also need to focus on appearing in AI searches on Gemini...
Sources: PayPal is attracting takeover interest from potential buyers after its stock fell ~46% over the last 12 months, giving PayPal a market value of ~$38.4B (Bloomberg)
Much has been made recently of the “Constitution” Anthropic published which will supposedly guide the development of its Claude LLM. Will be interested to see how those values are impacted once they start getting real, politically driven demands about what it says.
Microsoft Edge’s Secure Network feature is marketed as a VPN but only protects traffic within the Edge browser, not system-wide applications, according to a new report. It is an HTTP CONNECT proxy rather than a true VPN.
Why AI isn’t universally loved is a mystery, except for the part about the CEOs of AI companies incessantly talking about it eliminating most jobs, possibly destroying the world, etc. www.nytimes.com/2026/02/21/t...
📱How do you prefer to pay for electric car charging on the public network?
🫣 RFID? An app for every network? Apple Pay?
🤔 For new EV drivers, it can be confusing to know where to start.
📖 Check out our straightforward guide 👇
After a deadly raid, a Commie-Fascist FAKE AI power struggle erupts at the low IQ LOSER Pentagon - The Washington Post www.washingtonpost.com/technology/2...
The answer to keeping agentic AI secure is not to use it – let alone declare your OS as agentic from top to bottom, Microsoft. If you want to use it, then you'd better understand and properly implement privilege isolation, security segmentation, and all the other good things that you need when […]
New tutorial just dropped: 👀 how Google ADK works with Arize AX to power complex RAG flows with visibility for hallucination detection, retrieval quality, and answer-quality arize.com/blog/master...
I don’t think someone like Sam Altman is aware of how dangerous things can get for him as living standards get worse and people increasingly blame dystopian AI.
It genuinely wouldn’t surprise me if these Tech Bros start getting killed
Sources: Trump's "Board of Peace" is exploring a USD-pegged stablecoin for Gaza; a source says it will not be a "Gaza Coin" or a new Palestinian currency (Financial Times)
Honestly ... I don't think "vibe-porting" things from say C++ to Rust is a bad idea if your code base has tests that you can use to verify behaviour before and after a migration (without porting the tests themselves as well). Yes, the resulting code will initially be non-idiomatic, but that […]
Top AI firm alleges Chinese labs used 24K fake accounts to siphon US tech
NEWYou can now listen to Fox News articles! FIRST ON FOX: As Washington tightens export controls to preserve America’s artificial intelligence edge, top AI firm Anthropic says three China-based AI laboratories found another…
Despite launching with great fanfare, OpenAI’s ambitious Stargate data center project has been plagued by partner disagreements and missed goals, according to a new report from The Information.
🐞 Ladybird, the cross-platform, open source web browser, has begun rewriting core components from C++ to Rust, with the help of AI. The first port targets its JavaScript engine, leveraging memory safety and modern tools. alternativeto.net/news/2026/2...
Tomorrow (24 February) at 17:00, I'll be presenting on "Orchard Quantum Recoverability", a proposal that will improve Zcash's ability to cope with potential future quantum attacks. Luma sign-up link here: luma.com/jhu74jn3 (you will also need Google Meet).
Raspberry Pi CEO Eben Upton told me he'd be "a little bit surprised" if memory prices haven't leveled off by 2028. Budget phones and low-cost computers are just getting started feeling the pain.
The 16GB model of the Pi 5 has already nearly doubled in price.
Simon Willison@simon.fedi.simonwillison.net.ap.brid.gy
I've published the first two chapters of a new guide to Agentic Engineering Patterns - coding practices and patterns to help get the best results out of coding agents like Claude Code and OpenAI Codex
If you haven't already, feel free to share your experience on this post! Gained some traction to the Dev community and I would love to hear from you! #ai #cybersecurity #entrepreneur
Internal memo: Lightricks, creator of photo editing app Facetune, is splitting its consumer app unit from its GenAI unit LTX to better capture the growth in AI (Echo Wang/Reuters)
I worry that we don't have the economic, political, or cultural institutions in place to deal with rapid technological change. In fact, we just may have the worst political situation imaginable for it.
ANTHROPIC ALLEGES THREE CHINESE AI COMPANIES PROMPTED CLAUDE THROUGH FRAUDULENT ACCOUNTS - WSJ DEEPSEEK, MOONSHOT AI AND MINIMAX USED CLAUDE TO IMPROVE THEIR OWN PRODUCTS, ANTHROPIC ALLEGES THE METHOD CAN BE USED TO BUILD COMPETING PRODUCTS MORE QUICKLY AND CHEAPLY. ...
… Axios reported that the Pentagon had been considering cutting ties with Anthropic over the latter's insistence on retaining restrictions on how the U.S. military uses its models, which includes Claude AI. …
![Media from Sam[uel] Smith](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:6tmop7rnerqkwnd22jverhsz/bafkreiaq4oq4ik6vljiy3fkf2aowedvmx4a4bykm57zfxwlqthf7bpkx3i@jpeg)
![Media from Michael Georg Speller [THE GREY-APE]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:422uaueaquotbw7ie6th23qv/bafkreibf4cx6rckctfgtuq6bgged42zrhs4b6ktqyfva52iyrqrmqsuu6q@jpeg)
![Media from Michael Georg Speller [THE GREY-APE]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:422uaueaquotbw7ie6th23qv/bafkreiftusqs3pn22a2qcu7vt5iyiz6ydg5bs66djfgdiz26hvckdn6774@jpeg)
