Slashdot
News for Nerds. Stuff that Matters.
ChatGPT, Other Chatbots Approved For Official Use In the Senate
A senior Senate administrator has authorized the use of three AI chatbots—Google Gemini, OpenAI's ChatGPT, and Microsoft Copilot—for official work. The memo highlights Copilot’s ability to draft documents, summarize information, create talking points, and conduct research, while assuring that data stays within the secure Microsoft 365 Government environment. Usage guidelines remain decentralized, with individual Senate offices and committees setting their own rules. In contrast, the House has a more detailed policy that limits AI on sensitive data and deep‑fake creation.
Digg Relaunch Fails
After a brief relaunch as a Reddit‑style beta, Digg announced it is shutting down again, blaming an onslaught of AI‑generated bots and spam. CEO Justin Mezzell described a “hard reset,” noting that despite banning tens of thousands of accounts and...
Backblaze Hosts 314 Trillion Digits of Pi Online
Backblaze has partnered with StorageReview to make a 130 TB dataset containing 314 trillion digits of Pi publicly available. The original calculation generated about 2.1 PB of intermediate data, which was broken into roughly 200 GB chunks for easier download. The effort tested...
Atlassian CEO Cites AI Shift When Announcing Plan To Shed 1,600 Jobs
Atlassian announced it will eliminate 1,600 positions, about a tenth of its global workforce, as part of a strategic shift toward artificial intelligence. Founder Mike Cannon‑Brookes said AI is reshaping the skill mix required across the company, prompting the cuts....
Only Half of Americans Went To a Movie Theater In 2025, Study Finds
Only 53% of U.S. adults reported going to a movie theater in the past year, according to a Pew Research Center survey, while 7% have never attended. Ticket sales remain about 20% below pre‑pandemic levels, with 769.2 million tickets sold in...
Swiss E-Voting Pilot Can't Count 2,048 Ballots After USB Keys Fail To Decrypt Them
Swiss authorities suspended Basel‑Stadt's e‑voting pilot after 2,048 ballots could not be decrypted, despite three USB sticks containing the correct codes. The pilot, which served roughly 10,300 expatriates and 30 voters with disabilities, collected votes amounting to less than 4%...
Intel Demos Chip To Compute With Encrypted Data
Intel unveiled its Heracles processor at ISSCC, claiming up to a 5,000‑fold speed boost for fully homomorphic encryption (FHE) workloads compared with a top‑of‑the‑line Xeon server. The chip is fabricated on Intel’s 3‑nanometer FinFET process and occupies a die roughly...
Silicon Valley Is Buzzing About This New Idea: AI Compute As Compensation
Silicon Valley is adding AI inference compute as a fourth element of compensation, alongside salary, bonus and equity. Candidates now ask about dedicated GPU or token budgets for running models like Codex, reflecting growing scarcity of inference resources. Companies such...
Claude AI Finds Bugs In Microsoft CTO's 40-Year-Old Apple II Code
Microsoft Azure CTO Mark Russinovich demonstrated Claude Opus 4.6’s ability to reverse‑engineer his 1986 Apple II Enhancer utility, written in 6502 machine code. The AI decompiled the legacy program and uncovered several security flaws, including a silent incorrect‑behavior bug when a...
German Publishers Push Regulators To Fine Apple Over App Tracking Transparency
German publishers and advertising groups have asked the Bundeskartellamt to fine Apple over its App Tracking Transparency (ATT) system, claiming it unfairly blocks third‑party access to advertising data while Apple’s own apps remain exempt. Apple responded with proposals to use...
Qualcomm's New Arduino Ventuno Q Is an AI-Focused Computer Designed For Robotics
Qualcomm and Arduino have launched the Arduino Ventuno Q, an AI‑focused single‑board computer built for robotics and edge applications. The board is powered by Qualcomm’s Dragonwing IQ8 processor, featuring an 8‑core ARM Cortex CPU, an Adreno GPU and a Hexagon...
US Military Tested Device That May Be Tied To Havana Syndrome On Rats, Sheep
U.S. intelligence agents reportedly acquired a miniaturized microwave weapon from a Russian criminal network and secretly tested it on rats and sheep at a military facility. The animal experiments produced injuries that mirror the neurological symptoms reported by diplomats and...
Emails To Outlook.com Rejected By Faulty Or Overzealous Blocking Rules
Microsoft’s outbound email filters mistakenly rejected legitimate messages sent to Outlook.com, Live, Hotmail and MSN accounts last week. The over‑aggressive IP reputation checks generated 550 errors, blocking invoices, notifications and authentication emails despite senders having clean SNDS scores and being...
Stack Overflow Adds New Features (Including AI Assist), Rethinks 'Look and Feel'
Stack Overflow announced a beta redesign that introduces AI Assist, open‑ended questions, coding challenges, and an updated UI component library. The changes come as the platform’s monthly question volume has fallen 78% from its 2014 peak of 200,000 questions. The...
Duolingo Grows, But Users Disliked Increased Ads and Subscription Pushes. Stock Plummets Again
Duolingo reported a strong fourth‑quarter performance, with daily active users climbing to 52.7 million, paid subscribers up 28% to 12.2 million, and revenue reaching $282.9 million, marking a 35% year‑over‑year increase. The company also celebrated crossing the $1 billion annual revenue threshold for the...
AI Now Helps Manage 16% of America's Apartments
EliseAI, an AI housing startup, now handles leasing and management for roughly one in six U.S. apartments, about 16% of the market. The technology, accelerated by the COVID‑19 pandemic, enables self‑guided tours, instant text and voice interactions, and rapid response...
Has the AI Disruption Arrived - and Will It Just Make Software Cheaper and More Accessible?
Entrepreneur Paul Ford, co‑founder of AI‑driven platform Aboard, wrote a New York Times essay highlighting the rapid improvement of Anthropic’s Claude Code, now costing $200 per month and enabling massive code generation. He argues that AI coding can turn software...
HSBC To Investors: If India Couldn't Build an Enterprise Software Challenger, Neither Can AI
HSBC warned investors that AI‑generated code is unlikely to dethrone entrenched enterprise software giants. The bank cited India’s decades‑long experience customizing SAP and Oracle, noting that despite a massive engineering workforce, no home‑grown challenger has emerged. HSBC argues that success...
Discord Rival Maxes Out Hosting Capacity As Players Flee Age-Verification Crackdown
Discord’s global rollout of stringent age‑verification checks has sparked a wave of user backlash, prompting many gamers to abandon the platform for alternatives. TeamSpeak, a long‑standing voice‑chat service, has seen a surge in traffic, maxing out its hosting capacity across...
The First Signs of Burnout Are Coming From the People Who Embrace AI the Most
Harvard Business Review reports that embracing AI can inadvertently create burnout. UC Berkeley researchers spent eight months embedded in a 200‑person tech firm where workers voluntarily used AI tools. Without explicit pressure, employees expanded their to‑do lists, letting AI‑enabled efficiency...
ByteDance Suspends Seedance 2 Feature That Turns Facial Photos Into Personal Voices Over Potential Risks
ByteDance’s AI video platform Seedance 2.0 can ingest multiple media types, allowing users to blend up to nine images, three videos and three audio files into short clips. During testing, the tool’s facial‑to‑voice capability generated a voice nearly identical to...
OpenAI Starts Running Ads in ChatGPT
OpenAI has begun testing advertisements inside ChatGPT for logged‑in adult users on the Free and Go tiers in the United States. Ads are labeled as "sponsored," matched to conversation topics, and do not affect the model’s answers. Higher‑priced tiers such...
Discord Will Require a Face Scan or ID for Full Access Next Month
Discord will launch a global age‑verification system next month, automatically placing every user in a teen‑appropriate mode unless they prove they are adults. Unverified users will lose access to age‑restricted servers, stage channels, and will face stricter content filters and...
Carmakers Rush To Remove Chinese Code Under New US Rules
U.S. regulators are set to ban any Chinese‑origin software in cloud‑connected vehicle systems, forcing automakers to certify by March 17 that core code contains no Chinese provenance. The rule also covers advanced autonomous‑driving software and will expand to connectivity hardware by...
Automattic and the Internet Archive Team Up To Fight Link Rot
Automattic and the Internet Archive have launched a free, open‑source WordPress plugin called the Wayback Machine Link Fixer. The tool scans posts for broken outbound links, redirects visitors to archived Wayback Machine copies, and requests new snapshots when none exist....
Anthropic Launches Claude Opus 4.6 as Its AI Tools Rattle Software Markets
Anthropic unveiled Claude Opus 4.6, its most capable AI model yet, featuring a 1 million‑token context window and enhanced coding abilities. The upgrade promises better planning, longer autonomous tasks, and an internal debugging loop that catches errors. Independent GDPval‑AA testing shows Opus 4.6...
Say Hello To GoogleSQL
Google has retired the ZetaSQL name, rebranding its open‑source SQL analysis and parsing project as GoogleSQL. The change is purely nominal, with the underlying code, features, and team remaining unchanged. Internally, Google already referred to the dialect as GoogleSQL across...
'Vibe Coding Kills Open Source'
A trio of economists from Central European University, Bielefeld University and the Kiel Institute built a general‑equilibrium model showing that "vibe coding"—AI agents that automatically select, assemble and modify open‑source packages—breaks the traditional funding loop for open‑source projects. The model...
Amazon in Talks To Invest Up To $50 Billion in OpenAI
Amazon is reportedly negotiating a multi‑billion‑dollar investment of up to $50 billion in OpenAI, the creator of ChatGPT. The funding round could total $100 billion, potentially valuing OpenAI at $830 billion. Amazon CEO Andy Jassy is leading talks with OpenAI CEO Sam Altman,...
Backseat Software
Mike Swanson likens modern vehicle software to over‑zealous mobile apps, illustrating how frequent pop‑ups and forced tutorials interrupt drivers. He describes scenarios where a car repeatedly asks for feedback, pushes feature ads, and blocks essential gauges until the user complies....
Massive AI Chat App Leaked Millions of Users Private Conversations
Chat & Ask AI, a popular AI chatbot with over 50 million installs, suffered a massive data exposure due to a Firebase misconfiguration. An independent researcher accessed roughly 300 million messages belonging to more than 25 million users, revealing full conversation histories, timestamps,...
Software Company Bonds Drop As Investors' AI Worries Mount
Investors are pulling back from software company bonds as rapid AI advances raise doubts about business model stability. High‑profile moves like Google’s AI assistant launch and funding for Decagon AI have intensified fears, driving bond prices lower worldwide. While some...
OpenAI Releases Prism, a Claude Code-Like App For Scientific Research
OpenAI unveiled Prism, a free AI‑powered scientific research app built on the recently acquired Crixet LaTeX platform. The service replaces Crixet’s Chirp assistant with OpenAI’s GPT‑5.2 Thinking model, offering advanced LaTeX editing, automated bibliography generation, and content creation for teaching...
Gemini In Google Calendar Now Helps You Find the Best Meeting Time For All Attendees
Google has integrated its Gemini generative AI model into Calendar, introducing a "Suggested times" feature that automatically scans attendees' calendars to recommend optimal meeting slots based on availability, work hours, and conflicts. The tool also provides a one‑click rescheduling banner...
Anthropic Updates Claude's 'Constitution,' Just In Case Chatbot Has a Consciousness
Anthropic has published a revised 80‑page "Constitution" for its Claude chatbot, formalizing the ethical framework that guides its behavior. The living document defines four core values—broad safety, broad ethics, compliance with Anthropic guidelines, and genuine helpfulness—and includes a dedicated section...
When Two Years of Academic Work Vanished With a Single Click
Professor Marcel Bucher, a plant‑science faculty member at the University of Cologne, lost two years of research and teaching material after he temporarily disabled ChatGPT’s data‑consent setting. The action triggered an irreversible deletion of all his Plus‑account chats, prompts, and...
'Almost Everyone' Laid Off at Vimeo Following Bending Spoons Buyout
Vimeo has dismissed a large portion of its global workforce just months after Bending Spoons completed a $1.38 billion acquisition. The latest cuts, described by insiders as affecting "almost everyone," include the entire video engineering team and follow a 10 percent layoff...
Epic and Google Have a Secret $800 Million Unreal Engine and Services Deal
A federal judge disclosed an undisclosed six‑year, $800 million partnership between Epic Games and Google focused on Unreal Engine services and joint marketing. The agreement, kept largely confidential, involves Epic paying Google for cloud and AI services rather than co‑creating a...
EBay Bans Illicit Automated Shopping Amid Rapid Rise of AI Agents
eBay has revised its User Agreement to expressly forbid third‑party “buy‑for‑me” agents and large‑language‑model (LLM) chatbots from placing orders on its platform without explicit permission. The new clause, effective February 20 2026, expands the previous ban on generic bots to specifically target...
Workday CEO Calls Narrative That AI Is Killing Software 'Overblown'
Workday chief executive Carl Eschenbach told CNBC from Davos that the idea AI is destroying software businesses is "overblown" and outright false. He framed artificial intelligence as a tailwind, not a headwind, for Workday’s cloud‑based offerings. The comment comes as...
Autodesk To Cut 1,000 Jobs
Autodesk announced it will cut approximately 1,000 jobs, about 7% of its global workforce, as the final stage of a multi‑year restructuring. The layoffs focus primarily on customer‑facing sales roles, reflecting a shift toward a more digital, data‑driven selling model....
AI Company Eightfold Sued For Helping Companies Secretly Score Job Seekers
Eightfold AI, a venture‑backed hiring platform used by firms like Microsoft and PayPal, faces a California class‑action lawsuit alleging it scores job applicants without informing them. The suit claims the company violated the Fair Credit Reporting Act and state law...
56% of Companies Have Seen Zero Financial Return From AI Investments, PwC Survey Says
PwC’s Global CEO Survey of 4,454 executives across 95 countries reveals that 56% of companies have seen no financial benefit—neither higher revenue nor lower costs—from AI investments over the past year. Only 12% reported achieving both revenue growth and cost...
Hundreds Answer Europe's 'Public Call for Evidence' On an Open Digital Ecosystem Strategy
The European Commission launched a public call for evidence on open digital ecosystems, running from 6 January to 3 February 2026. More than 370 submissions have already been received, reflecting strong stakeholder interest. The evidence will shape a Commission communication that outlines concrete...
AI Has Made Salesforce Engineers More Productive, So the Company Has Stopped Hiring Them, CEO Says
Salesforce CEO Marc Benioff announced that the company’s engineering headcount has stayed mostly flat over the past year as internal AI tools dramatically boost developer productivity. He said the firm now has about 15,000 engineers who are "more productive than...
Ruby on Rails Creator Says AI Coding Tools Still Can't Match Most Junior Programmers
Ruby on Rails creator David Heinemeier Hansson told a podcast that current AI coding assistants still fall short of the output produced by most junior developers. At 37 Signals, 95% of the code for their Kanban product Fizzy was written by...
Oracle Trying To Lure Workers To Nashville For New 'Global' HQ
Oracle is positioning Nashville as its new global headquarters, pledging a massive riverfront campus and a 8,500‑job creation target by 2031. The company is offering tens of thousands of dollars in relocation incentives to attract cloud and AI talent, making...
Beijing Tells Chinese Firms To Stop Using US and Israeli Cybersecurity Software
Beijing has ordered Chinese companies to cease using cybersecurity software from roughly a dozen U.S. and Israeli vendors, citing national security concerns. The directive targets products from Broadcom‑owned VMware, Palo Alto Networks, Fortinet, and Israeli firm Check Point. The move...
Bezos's Vision of Rented Cloud PCs Looks Less Far-Fetched
Jeff Bezos likened today’s desktop towers to obsolete generators, predicting a future where users rent cloud‑based PCs from providers like AWS and Azure. Recent market pressures—soaring DRAM costs, Micron’s exit from consumer memory, and tightening SSD supply—are making that vision...
UK Police Blame Microsoft Copilot for Intelligence Mistake
West Midlands Police admitted that Microsoft Copilot fabricated a non‑existent football match between West Ham and Maccabi Tel Aviv, which was then embedded in an intelligence report. The erroneous detail prompted the ban of Israeli supporters at a subsequent game....
