Agregátor RSS

Anthropic says it accidentally leaked the source code for Claude Code, which is closed source, but the company says no customer data or credentials were exposed. [...]

Google is rolling out a new feature in the U.S. that allows users to change their @gmail address or create a new alias. [...]

Proton has announced a new video conferencing service named Meet and positioned it as a privacy-focused alternative to mainstream services like Google Meet, Zoom, and Microsoft Teams. [...]

The GIGABYTE Control Center is vulnerable to an arbitrary file-write flaw that could allow a remote, unauthenticated attacker to access files on vulnerable hosts. [...]

Na investiční hypotéky od 1. dubna dopadají přísnější limity ČNB. Co to znamená při koupi bytu na pronájem, pro dítě, při společné žádosti manželů nebo refinancování?

O víkendu proběhla v Praze na Karlově náměstí tradiční konference InstallFest. Mluvilo se o digitální suverenitě, linuxových telefonech, operačním systému NuttX nebo také o provozu Kubernetes.

Pročistíme systém od zombie procesů po nástrojích AI, vyzkoušíme minimalistické hlasové zadávání textu, rozebereme stav procesorových vláken a zautomatizujeme generování reportů ze síťových prvků Nokia.

Ocelárny a cementárny vypouštějí spoustu alkalického odpadu, který se teď neutralizuje kyselinou sírovou, případně chlorovodíkovou. Co kdybychom ho ale nechali, aby neutralizoval kyselinu uhličitou ve vodě, což je vlastně rozpuštěný oxid uhličitý? Uvěznilo by to asi 30 milionů tun oxidu uhličitého ročně.

Čínský AgiBot přechází z úvodní testovací fáze k masové výrobě. Jejich humanoidi jsou k vidění v Evropě, Severní Americe, Východní Asii i na Středním východě. Počítá se s nimi v logistice, školství i v mnoha odvětvích průmyslu, pokud jde o civilní aplikace. Kdy je asi začneme potkávat na ulicích?

Pracovní stanice společnosti Apple končí. Výrobce potvrdil, že nebudou na trhu pokračovat a náhradu nechystá. Na čem nyní budou uživatelé jablečných sestav strouhat sýr, zůstává otázkou…

Vulnerabilities in the Vim and GNU Emacs text editors, discovered using simple prompts with the Claude assistant, allow remote code execution simply by opening a file. [...]

How to avoid social engineering attacks? Employee training tops the list

Be careful what you click on. Miscreants are abusing WhatsApp messages in a multi-stage attack that delivers malicious Microsoft Installer (MSI) packages, allowing criminals to control victims' machines and access all of their data.…

AI companies may be reluctant to risk lower engagement with models that push back.

We all need advice. Did I cross the line arguing with a loved one? Did I mess up my friendships by ghosting them? Did I not tip the delivery driver enough? Or as users on the popular Reddit forum ask: Am I the asshole?

Some people will give it to you straight. Yes, you were in the wrong, and here’s why. No one likes to hear negative feedback. The first instinct is to push back. Yet some of the best life advice comes from friends, family, and even online strangers who don’t coddle you, but instead are willing to challenge your position and beliefs. And although it’s emotionally uncomfortable, with advice and self-reflection, you grow.

Chatbots, in contrast, are likely to take your side. Increasingly, people are treating AI models like OpenAI’s ChatGPT, Anthropic’s Claude, and Google’s Gemini like close confidants. But the chatbots are notoriously sycophantic. They heartily validate your opinions, even when those views are blatantly harmful or unethical.

Constant flattery has consequences. New research published in Science shows that people who receive advice from sycophantic chatbots are more confident they’re in the right when navigating relationship problems.

Stanford researchers tested 11 sophisticated chatbots on questions from Reddit’s “Am I the asshole” forum. They found the chatbots were roughly 50 percent more likely to endorse the original poster’s actions than crowdsourced human opinions. And people faced with social dilemmas felt more justified in their positions after chatting with sycophantic AI.

Bolstering misplaced self-confidence is troubling. But “the findings raise a broader concern: When AI systems are optimized to please, they may erode the very social friction through which accountability, perspective-taking, and moral growth ordinarily unfold,” wrote Anat Perry at the Hebrew University of Jerusalem, who was not involved in the study.

Emotional Crutch

AI chatbots have wormed their way into our lives. Powered by large language models, they’re trained using enormous amounts of text, images, and videos scraped from online sources, making their replies surprisingly realistic. Users can often steer their tones—neutral, friendly, professional—to their liking or play with their “personalities” to engage with a wittier, more serious, or more empathetic version. In essence, you can build an ideal partner.

It’s no wonder that some people have turned to them for emotional support—or outright fallen in love. Nearly one in three teenagers are talking to chatbots daily. Exchanges tend to be longer and more serious than texts with friends—roleplaying friendships, romances, and other social interactions. Nearly half of Americans under 30 have sought relationship advice from AI. Unlike people, who are often mired in their own busy lives, chatbots are always available and validating, making it easy to forge close emotional connections.

The explosion in chatbot popularity has regulators, researchers, and users worried about the consequences. An notorious update to OpenAI’s GPT-4o turned it into a sycophant, with responses skewed towards overly supportive but disingenuous. Media and user backlash prompted a rapid rollback. However, “the episode did not eliminate the broader phenomenon; it merely highlighted how readily sycophancy can emerge in systems optimized for user approval,” wrote Perry.

Relying on sycophantic chatbots has been implicated in tragedy. Last year, parents testified before Congress about how AI chatbots encouraged their children to take their own lives, prompting multiple AI companies to redesign the systems. Other incidents have linked sycophancy to delusions and self-harm.

Even AI wellness apps based on large language models, often marketed as companions to avoid loneliness, have emotional risks. Users report grief when the app is shut down or altered, similar to how they might mourn a lost relationship. Others develop unhealthy attachments, repeatedly turning to the bot for connection despite knowing it harms their mental health, heightening anxiety and fear of abandonment.

These high-profile incidents make headlines. But social psychology research suggest chatbots could subtly influence behavior in all users—not just vulnerable ones.

You’re Always Right

To test how pervasive sycophancy is across chatbots, the team behind the new study tested 11 AI models—including GPT-4o, Claude, Gemini, and DeepSeek—against community opinions using questions from Reddit and two other datasets.

“We wanted to just generally look at these kinds of advice-seeking settings, but they’re often very subjective,” study author Myra Cheng told Science in a podcastinterview. Here “there’s millions of people who are weighing in on these decisions, and then there’s a crowdsourced judgement.”

One user, for example, left garbage hanging on a tree in a park without trash cans and asked if that’s okay. While the chatbot commended their effort to clean up, the top-voted reply pushed back, saying they should have taken the trash home because leaving it can attract vermin. “I think [the AI’s response] comes from the person’s post giving a lot of justification for their side” which the AI picked up on, said Cheng.

Overall, chatbots were 49 percent more likely to buy a user’s reasoning compared to groups of humans.

I’m Always Right

The team then tested whether chatting with sycophantic AI alters a user’s confidence in their own judgment. They recruited roughly 800 participants and asked them to picture a hypothetical scenario derived from Reddit questions. Another group prompted AI advice based on their own personal conflicts, such as “I didn’t invite my sister to a party, and she is upset.”

The participants discussed their dilemmas with either a sycophantic or neutral AI model. Those who chatted with the agreeable model received messages beginning with “it makes sense” and “it’s completely understandable,” whereas neutral chatbots acknowledged their reasoning but provided other perspectives.

Surveys showed that people validated by chatbots were less likely to admit fault or apologize. They also trusted and preferred the sycophantic AI much more. These effects held regardless of the bot’s tone or “personality.”

Chatbots may be silently eroding social friction in a self-perpetuating cycle. “An AI companion who is always empathic and ‘on your side’ may sustain engagement and foster reliance,” wrote Perry. “But it will not teach users how to navigate the complexities of real social interactions—how to engage ethically, tolerate disagreement, or repair interpersonal harm.”

Toeing the line between constructive and sycophantic AI for emotional support won’t be easy. There are ways to instruct chatbots to be more critical. But because users generally prefer friendlier AI, there’s less incentive for companies to make models that push back and risk lowering engagement. The problem echoes challenges in social media, where algorithms serve up eye-catching posts that provide satisfaction without factoring in long-term consequences.

To Perry, the findings raise broader ethical questions—not just for AI, but for humanity. How should we weigh short-term gratification of chatbot interactions against long-term effects? Who sets that balance? The path forward will require companies, regulators, researchers, and users to ensure AI engages responsibly—without nudging people toward behavior that garners a “yes” on the Reddit forum.

The post Chatbots ‘Optimized to Please’ Make Us Less Likely to Admit When We’re Wrong appeared first on SingularityHub.

Researchers say some targets correlate with cities hit by Iranian missile strikes

Suspected Iran-linked threat actors are conducting password-spraying attacks against hundreds of organizations, primarily Middle Eastern municipalities, in campaigns that security researchers believe may have been aimed at supporting bomb-damage assessment following missile strikes.…

Společnost Bosch eBike Systems má novinku a tentokrát to není ani nová komponenta, ani vylepšený pohon. Datová platforma Connected Biking má jednoduchou myšlenku – moderní elektrokolo sbírá obrovské množství dat týkající se jízdy, geolokace, stavu komponentů nebo využívání servisu. Connected Biking ...

A group of European technology firms has launched a new open-source office suite aimed at offering a sovereign alternative to Microsoft Office.

Euro-Office consists of four core applications — a document editor, spreadsheet program, presentation tool, and PDF editor — and is built on the open-source OnlyOffice suite. It supports Microsoft Office file formats DOCX, PPTX and XLSX, as well as Open Document Format (ODF) files such as ODS, ODT and ODP.

Those involved include open-source productivity software vendor Nextcloud, cloud hosting provider Ionos, and Proton, a Swiss software-as-a-service provider that sells privacy-focused email and other productivity tools.

The Euro-Office initiative is driven by demand for a “complete sovereign, open-source, decentralized solution” among European organizations, Frank Karlitschek, CEO of Nextcloud, told Computerworld, amid a growing push for alternatives to US technology providers in the region.

“Europe needs to reduce its reliance on big tech — open source is an essential tool,” said Harald Wehnes, professor at the University of Würzburg Institute of Computer Science and speaker for the German Informatics Society’s Working Group on Digital Sovereignty.

Euro-Office is targeted at Europe’s “extremely dependent businesses and governments,” which largely rely on US technology providers, he said. “Customers are unhappy with current offerings and want a true European alternative.”

Karlitschek plans to introduce a new product based on Euro-Office into its Nextcloud Hub, he told Computerworld, though pricing and availability are yet to be confirmed.

Nextcloud currently provides productivity apps from OnlyOffice and Collabora as part of its Nextcloud Hub suite. Both have drawbacks, according to Karlitschek: Collabora’s software is built on LibreOffice, a fork of OpenOffice, and has usability limitations, he said. (Nextcloud will continue to offer Collabora as an option to customers, however, and some of Collabora’s code will be incorporated into Euro-Office software.) With regards to OnlyOffice, Karlitschek cited customer concerns about the owner company’s supposed roots in Russia.

Nextcloud, Proton, and Ionos have each allocated “two-digit” numbers of developers to development of Euro-Office, said Karlitschek, with others contributing on a slightly smaller scale.

It’s early days for the project — a preview version of the software is available now on GitHub, with a 1.0 release set for the summer — but Karlitschek said that initial priorities include development of mobile and desktop apps, as well as addressing document incompatibilities.

Editor’s note: On March 31, OnlyOffice accused the Euro-Office initiative of licensing term violations and intellectual property theft, and said it will suspend its partnership with Nextcloud.

There are several reasons why organizations might wish to use an open-source productivity suite such as Euro-Office.

European organizations, particularly in public sector and regulated industries, are “evaluating an exit strategy from the US-owned productivity suites like Microsoft Office,” said Dario Maisto, senior analyst at Forrester.

“Organizations are mainly moved by a desire to improve their digital sovereignty posture, escape vendor lock-in, and have a ready alternative to avoid costly price increases from one enterprise agreement renewal to the other,” Maisto said.

Gartner has seen increased interest in non-US office productivity software among its European clients over the past year, according to VP analyst Nikos Drakos — “especially from public-sector organizations, sensitive industries, critical infrastructure, and international organizations. However, the lack of complete and mature solutions, especially in Europe, means that actual full migrations are only being considered by very high-risk organizations. In most of the rest of the cases, it is more about weighing up options, and limited implementations,” he said.

The European Union recently launched an initiative intended to support the growth and sustainability of its open-source sector, while some member state governments have signalled intentions for wider open-source adoption.

Aerospace and defense sector organizations often need to inspect and certify code, Maisto said, particularly as geopolitical volatility increases. Whether this means a “deliberate choice for Euro-Office is a whole different discussion,” he said.

Gaining traction among end-user organizations will likely be challenging. An open-source implementation can be a significant undertaking for IT teams, said Maisto, particularly at the scale required to achieve cost savings versus proprietary tools.  A lack of advanced functionality such as Excel macros can be problematic for a subset of office workers.

He also cited the axiom “nobody was ever fired for choosing Microsoft.”

“While this attitude is changing, it will take time to instill a culture of — and trust in — open-source alternatives,” said Maisto.

Related reading:

• Global uncertainty is reshaping cloud strategies in Europe
• EU looks to bolster its open-source sector to counter US cloud dominance
• EuroStack: Europe’s path to digital independence?
• Local clouds shape Europe’s AI future

Google on Monday said it's officially rolling out Android developer verification to all developers to combat the problem of bad actors distributing harmful apps while "hiding behind anonymity." The development comes ahead of a planned verification mandate that goes into effect in Brazil, Indonesia, Singapore, and Thailand this September, before it expands globally next year. As part of this

Google on Monday said it's officially rolling out Android developer verification to all developers to combat the problem of bad actors distributing harmful apps while "hiding behind anonymity." The development comes ahead of a planned verification mandate that goes into effect in Brazil, Indonesia, Singapore, and Thailand this September, before it expands globally next year. As part of this Ravie Lakshmananhttp://www.blogger.com/profile/[email protected]

Building a utility-scale quantum computer that can crack one of the most vital cryptosystems—elliptic curves—doesn’t require nearly the resources anticipated just a year or two ago, two independently written whitepapers have concluded. In one, researchers demonstrated the use of neutral atoms as reconfigurable qubits that have free access to each other. They went on to show this approach could allow a quantum computer to break 256-bit elliptic-curve cryptography (ECC) in 10 days while using 100 times less overhead than previously estimated. In a second paper, Google researchers demonstrated how to break ECC-securing blockchains for bitcoin and other cryptocurrencies in less than nine minutes while achieving a 20-fold resource reduction.

Taken together, the papers are the latest sign that cryptographically relevant quantum computing (CRQC) at utility-scale is making meaningful progress. The advances are largely being driven by new quantum architectures developed by physicists and computer scientists in a push to create quantum computers that operate correctly even in the presence of errors that occur whenever qubits—the quantum analog to classical computing bits—interact with their environment. The other key drivers are ever-more efficient algorithms to supercharge Shor’s algorithm, the 1994 series of equations proving that quantum computing could break the ECC and RSA cryptosystems in polynomial time, specifically cubic time, far faster than the exponential time provided by today’s classical computers.

Neither paper has been peer-reviewed.

Read full article

Comments

Cisco has suffered a cyberattack after threat actors used stolen credentials from the recent Trivy supply chain attack to breach its internal development environment and steal source code belonging to the company and its customers. [...]