Category: Biz Tech

TypeSafe AI has released Jev, a transformer-based artificial intelligence model designed to make structured decisions for software rather than generate conversational text, drawing strong early interest from developers seeking faster and cheaper automation tools. The San Francisco start-up, founded by former OpenAI researcher Diogo Almeida, introduced Jev as its first “System One” model after two years in stealth. Almeida helped develop research behind ChatGPT and reinforcement learning from human feedback, a training approach that helped make large language models better at

Software stocks are staging a sharp recovery as stronger-than-expected earnings and fresh calls for restraint in advanced artificial intelligence development ease fears that AI will quickly undermine the sector’s business models. The iShares Expanded Tech-Software Sector ETF has climbed about 17 per cent since the end of the first half, reversing part of a 14 per cent decline during the opening six months of 2026, when investors heavily favoured chipmakers and other AI infrastructure plays. Over the same period, the Philadelphia

Cybersecurity researchers have identified a new Android malware strain that uses generative artificial intelligence to help attackers navigate infected devices and steal banking credentials, payment information and authentication codes. Zimperium’s zLabs team disclosed the malware, named RatHat, on September 16, saying its technical indicators point to threat actors who appear to be operating from China. The researchers stopped short of definitively attributing the campaign, basing the assessment chiefly on Mandarin-language prompts embedded in the malware’s AI-driven automation system. RatHat combines spyware, remote-access

OpenAI has asked a federal judge in California to dismiss lawsuits brought by survivors and families affected by the Tumbler Ridge mass shooting, arguing that British Columbia is the appropriate forum for claims arising from the February attack. Motions filed in the US District Court for the Northern District of California invoke the doctrine of forum non conveniens, which allows a court to decline a case when another available forum is substantially more suitable. OpenAI says British Columbia is both available

Software teams building cloud-native, edge and AI-driven systems are increasingly evaluating NATS as an alternative to RabbitMQ, drawn by lower-latency messaging, simpler deployment and a communication model designed around distributed services rather than conventional broker-centric queues. The shift has gained fresh attention with NATS Server 2.15, released on September 17, which concentrates on reliability and operations. Its maintainers reworked stream and consumer scaling, movement and peer changes, improved backups and source recovery, and reported substantial performance gains for replicated JetStream streams

SpaceX has held informal discussions about buying customer and operational data from troubled or defunct startups to train artificial intelligence models, people familiar with the talks say, as the company looks for lower-cost information beyond material available on the public internet. The discussions are taking place within SpaceXAI, the artificial intelligence operation created after SpaceX acquired Elon Musk’s xAI business in February. No transaction has been announced, the talks are preliminary and may not lead to any purchases. SpaceX has not

Dubai is set to gain a mixed-use development where robots are planned to handle reception, security, cleaning, deliveries, coffee service and selected wellness functions, as Fakhruddin Properties advances a Dh700 million project in Dubai Land Residence Complex. The development, called Tréppan Vision, was unveiled at IPS 2026 and is scheduled for completion in the fourth quarter of 2029. Fakhruddin Properties says the scheme is designed to extend automation beyond individual apartments by operating specialised robotic systems throughout the building,

OpenAI chief Sam Altman, Anthropic chief Dario Amodei and xAI owner Elon Musk have publicly backed a slower pace for frontier artificial intelligence development, an unusual alignment among fierce rivals as safety concerns intensify and Altman says OpenAI will not pursue a stock-market listing in 2026. The convergence followed an essay by Amodei calling for the industry to “pace the frontier” rather than continue scaling advanced models as quickly as possible. He proposed permanent access for independent safety evaluators inside leading

Security specialists from the United States and China have proposed a set of nuclear-style safeguards for military artificial intelligence, including red lines around nuclear command systems, mandatory human control over major cyber operations and a dedicated bilateral crisis hotline. The proposals were published on September 9 by Melanie Sisson, a senior fellow at the Brookings Institution, and Tianjiao Jiang, an associate professor at Fudan University. Both participate in a US-China dialogue on artificial intelligence and national security convened by Brookings and

Salesforce has launched AIforce, a new interface layer designed to let workers and artificial intelligence agents use live customer relationship management data, workflows and permissions from tools including Claude, Slack and Amazon Quick without opening the Salesforce application. Unveiled at Dreamforce 2026, AIforce exposes Salesforce data, business logic, semantic context, actions and governance through open interfaces while retaining the access controls already configured by customers. Salesforce said requests continue to run through existing permissions and business rules, so users and agents

The UAE has highlighted an artificial intelligence system developed by GenBio AI with contributions from Mohamed bin Zayed University of Artificial Intelligence researchers that can simulate human cellular responses to drugs and other interventions. The UAE Embassy in Washington drew attention on Wednesday to AIDO Cell, a virtual-cell “world model” designed to represent biological activity across multiple levels, from DNA and RNA through proteins and whole-cell behaviour. GenBio AI unveiled a preview of the system in August and says it could

A US federal judge has ordered Elon Musk’s X Corp and SpaceXAI to disclose any agreement behind their resolution of antitrust claims against Apple, even as the companies press ahead with the same lawsuit against OpenAI. Judge Mark Pittman of the US District Court for the Northern District of Texas directed X and SpaceXAI to provide the court with any agreement or combination of agreements with Apple relating to the resolution, after OpenAI sought access to the terms. The order followed

Meta Platforms is preparing a camera-free pair of smart glasses, internally code-named Luna, for an expected autumn debut, with shipments targeted for October, according to people familiar with the company’s plans. The glasses are designed around voice-based artificial intelligence rather than photography or video, marking a notable departure from Meta’s existing camera-equipped eyewear. People familiar with the product say Luna will contain six microphones, open-ear speakers and a physical button on one temple for quickly activating Meta AI. The device could be

Cybersecurity shares surged to the top of the S&P 500 on Monday as investors shifted towards digital-defence companies after fresh warnings about the risks of rapidly advancing artificial intelligence unsettled the broader technology trade. CrowdStrike jumped 13.8 per cent to a record closing high of $235.38, while Palo Alto Networks gained 13.1 per cent and Fortinet rose about 9 per cent. The gains stood out in a session when the S&P 500 fell 0.5 per cent, the Nasdaq Composite dropped 0.6

SUSE has enabled its enterprise real-time Linux platform for LF Energy’s SEAPATH architecture, giving electricity grid operators and equipment vendors a commercially supported operating system option for virtualised digital substations. The integration, developed with open-source engineering specialist Savoir-faire Linux, combines SUSE Linux Enterprise Real Time with Kernel-based Virtual Machine, or KVM, virtualisation and cluster support. SUSE said the work makes it the first enterprise Linux vendor to provide validated real-time operating system support for the open-source SEAPATH architecture. SEAPATH is designed as

Microsoft has released an out-of-band Windows update to correct several faults introduced by its September security patches, including Remote Desktop instability, Hyper-V Linux file-sharing failures and problems affecting some USB audio devices. The emergency servicing release, issued on September 14, applies across multiple supported Windows editions, with separate packages for Windows 11, Windows 10 and Windows Server versions. Microsoft said the updates address problems traced to security fixes distributed on September 8, which disrupted functions used by businesses, developers and administrators. One

OpenAI is using hundreds of outside contractors to examine real ChatGPT conversations and rate the system’s replies, with some reviewers earning more than $50 an hour, according to leaked internal materials that have sharpened scrutiny of the company’s privacy practices. The work is described in instruction guides, Slack messages and reviewer materials under an internal initiative known as Project Lily. Contractors are shown user prompts and, in some cases, complete exchanges with ChatGPT, then asked to summarise the user’s intent, assess

Netflix, Amazon and YouTube have created a Washington-based lobbying coalition to press for policies favouring consumer choice and technology-neutral treatment of streaming services as political scrutiny intensifies over the migration of live sport from traditional television. The Streaming Access and Choice Alliance, or SACA, was announced on Monday with the three companies as founding members and TechNet, a technology industry advocacy group, leading the organisation. Mike Ward, TechNet’s senior vice-president of federal policy and government relations, will head the coalition. SACA said

President Donald Trump has rejected calls for tougher artificial intelligence safeguards, arguing that existing US powers are sufficient even as security specialists warn that faster AI systems are widening the attack surface for cryptocurrency platforms and decentralised finance. Trump said on September 14 that additional safeguards were unnecessary and dismissed warnings from technology executives about uncontrolled development. His position places the White House at odds with industry figures who have urged stronger testing, independent audits and tighter controls on advanced models

Cybersecurity stocks jumped on Wall Street on Monday as warnings over the pace and control of increasingly autonomous artificial intelligence systems drove investors away from semiconductor shares and towards companies expected to benefit from heavier spending on digital defence. CrowdStrike closed about 13.9% higher and Palo Alto Networks gained roughly 13.1%, placing both among the strongest performers in the S&P 500. Intraday gains approached 15% for CrowdStrike and 14% for Palo Alto Networks, while Nvidia, AMD and other AI-linked chipmakers fell

Calls by leading artificial intelligence executives to slow development of advanced models pressured Asian semiconductor shares on Monday, but investors and strategists said robust spending on computing infrastructure was likely to keep the longer-term AI investment cycle intact. Samsung Electronics fell as much as 4.1% in early trading, while South Korea’s SK Hynix dropped 5.8%. Japan’s Advantest declined 4.7% and Kioxia Holdings slid 9.3% as markets assessed whether tighter safety controls and a slower model-development pace could weaken demand for high-end

Artificial intelligence leaders are facing mounting resistance from investors and President Donald Trump after calls by Anthropic and OpenAI to slow the advance of frontier models triggered sharp losses in technology shares and exposed a widening dispute over how quickly the industry should proceed. Anthropic chief executive Dario Amodei has urged leading developers to reduce the pace at which they improve model capabilities, arguing that safety systems and independent oversight need more time to catch up. OpenAI chief executive Sam Altman

Z. AI Co. has raised about $5 billion through a Hong Kong share placement and convertible bond sale, returning to the market within days of a 60-day issuance lock-up ending as heavy spending on artificial intelligence models continues to consume cash. The Beijing-based company, formerly known as Zhipu AI, sold 21.97 million new H shares at HK$714 each, raising roughly HK$15.7 billion, or about $2 billion. It also issued 20.14 billion yuan, or about $3 billion, of zero-coupon convertible bonds due

China’s state security minister Chen Yixin has warned that artificial intelligence poses six major risks to national security, ranging from political manipulation and cyberattacks to data leakage, social disruption and military transformation. Chen, who heads the Ministry of State Security, set out the risks in an article published on Sunday by China Cyberspace magazine, arguing that Beijing must build a stronger system for monitoring, regulating and responding to dangers created by rapidly advancing AI. His intervention came as Anthropic chief executive Dario

Dmitry Medvedev has accused leading artificial intelligence companies of overstating fears that their technology could destroy humanity, arguing that commercial self-interest rather than existential concern is shaping their public warnings. Russia’s Security Council deputy chairman said the creators of generative AI systems were “not being sincere” when presenting themselves as alarmed by the possibility that advanced models could become an existential threat. He linked that scepticism to what he described as corporate greed, suggesting companies would not willingly abandon technologies capable

Kubernetes 1.37 has expanded the container orchestration platform’s security, autoscaling and workload-scheduling capabilities, with several long-running features reaching production maturity and new controls aimed at large enterprise and artificial-intelligence clusters. The release, named Garhwal and issued on August 26, contains 67 enhancements: 16 graduating to stable status, 23 moving to Beta, 27 entering Alpha and one involving deprecation or removal. The changes span workload identity, admission control, autoscaling, storage, device management and scheduling. A significant security change is the graduation of Pod

Revolut has disclosed that a limited number of customers had sensitive personal and financial information exposed after the fintech company responded to fraudulent requests sent from an unauthorised account operating within a legitimate government agency’s email domain. The company said the requests appeared authentic because they came from an official government domain and carried valid domain authentication credentials. Revolut later determined that the email account had been created or controlled without authorisation and had been used to obtain customer information under

OpenAI will not pursue an initial public offering this year, chief executive Sam Altman has said, arguing that mounting questions over artificial intelligence safety make 2026 the wrong moment to take the company public. Altman said OpenAI did not feel pressure to proceed with a flotation and indicated that safety and alignment work should take priority over the demands that would come with public-market ownership. “Given everything happening with safety, right now would be an ill-advised moment to go public,” he

Anthropic chief executive Dario Amodei has called for the pace of frontier artificial intelligence development to slow, drawing public backing from OpenAI’s Sam Altman and Elon Musk as developers confront concerns about systems helping to create more capable successors. Amodei said in an essay published on Saturday that AI companies should “pace the frontier” rather than continue maximising capability gains while safeguards lag behind. He argued that progress could remain rapid, but developers should deliberately create more time for alignment research,

Anthropic says its safety systems blocked a scientist’s request for Claude to help write a grant application for research that could have made the chikungunya virus more transmissible and better able to evade immunity, prompting a wider investigation into potentially dangerous biological work linked to military and state-backed institutions. The artificial intelligence company disclosed the case in a September threat-intelligence report describing five instances in which researchers used, or attempted to use, Claude for work that could support biological weapons development.

OpenAI has temporarily stopped new subscriptions and upgrades to its $200-a-month ChatGPT Pro tier as demand for its new GPT-6 Astra model strains computing capacity. The restriction, introduced on September 10, applies to new sign-ups for the Pro $200 plan, also called Pro 20X, and to users seeking to move to it from Free, Go, Plus or the lower-priced Pro $100 tier. Existing Pro $200 subscribers can continue using their accounts and their subscriptions will renew normally, OpenAI said in updated

Adobe posted record third-quarter revenue and crossed one billion monthly active users, but investors remained cautious as its expanding freemium strategy slowed some recurring-revenue indicators and tempered near-term growth expectations. The software group reported revenue of $6.76 billion for the fiscal third quarter ended August 28, up 13% from a year earlier and above market expectations of about $6.7 billion. Adjusted earnings were $6.13 a share, also ahead of forecasts, while net income rose to $1.83 billion, or $4.62 a diluted

Security researchers have mapped a pay-per-install malware operation using YouTube gaming channels and poisoned search results to distribute more than 10,000 distinct samples of a custom loader, exposing a cybercrime delivery network operating at substantial scale. Palo Alto Networks’ Unit 42 said the activity, tracked as CL-CRI-1171, funnels victims through two main routes: gaming-related YouTube content and search-engine optimisation poisoning that steers users towards trojanised software downloads. Both routes ultimately lead to the same pay-per-install infrastructure, where OfferLoader installs payloads chosen

OpenAI has temporarily stopped new sign-ups and upgrades to its $200-a-month ChatGPT Pro plan after demand for its GPT-6 Astra model put heavy pressure on computing capacity. The company confirmed that the restriction took effect on September 10 and applies specifically to the Pro $200 tier, also described in its support material as Pro 20X. Existing subscribers on that plan remain unaffected, while the lower-priced Pro $100 option can still be purchased or renewed. Thibault Sottiaux, an OpenAI product leader overseeing ChatGPT