Category: Biz Tech

OpenAI has moved to shut down its experimental social media platform Sora following mounting scrutiny over the misuse of AI-generated video, including the creation of non-consensual content and increasingly convincing deepfakes that alarmed regulators, creatives and digital safety advocates.The decision comes after months of debate surrounding the rapid adoption of generative video tools, with Sora gaining traction for its ability to produce short-form clips from simple text prompts. What began as a showcase of technical progress quickly evolved into a

Semiconductor chips have moved beyond their traditional role as hardware components to become strategic gateways shaping the trajectory of the global artificial intelligence economy, according to legal and technology analysts tracking the sector’s rapid evolution.At the centre of this shift stands Nvidia, which continues to command a dominant share of the market for graphics processing units used in AI training. Its high-performance chips, particularly those tailored for large-scale machine learning workloads, have become foundational infrastructure for companies building advanced AI

Cybercriminals are exploiting the popularity of artificial intelligence tools by distributing fake ChatGPT-themed Android applications designed to steal user credentials and compromise social media accounts, according to cybersecurity researchers tracking a coordinated phishing campaign.The operation centres on abuse of Google’s Firebase App Distribution platform, a legitimate service used by developers to test applications before public release. Attackers are leveraging the platform to send invitation-style emails that mimic official app testing notifications, prompting recipients to download what appear to be pre-release

Elon Musk has unveiled a plan to integrate Tesla, SpaceX and xAI into a new semiconductor initiative in Texas, marking a deeper push into controlling the hardware underpinning artificial intelligence systems. The project, described internally as “Terafab”, aims to develop advanced chips tailored for AI training, robotics and space applications.The initiative reflects a broader shift among technology companies seeking to reduce dependence on external suppliers such as Nvidia and Taiwan Semiconductor Manufacturing Company, whose dominance in AI chips has tightened

Amazon is developing an artificial intelligence-driven smartphone aimed at revitalising its voice assistant ecosystem, marking a strategic return to hardware after the commercial failure of its Fire Phone. The project, internally referred to as “Transformer”, is understood to centre on embedding generative AI deeply into the device’s operating system, signalling a shift from voice-first assistants to context-aware digital agents.The initiative comes at a time when large technology firms are racing to integrate advanced AI into consumer devices, transforming smartphones into

Messaging platform Telegram has fully repaid bonds issued five years ago, marking a significant financial milestone as the company signals sustained profitability and stronger investor confidence, according to a statement from its founder Pavel Durov.Durov confirmed that the debt raised in 2021 has now been cleared in full, underscoring a turnaround in the platform’s financial trajectory. The repayment follows a period during which Telegram introduced a series of monetisation initiatives that reshaped its revenue model, culminating in profitability achieved during

Microsoft is moving to scale back several artificial intelligence-driven features in Windows 11 while prioritising performance improvements, a shift that reflects mounting user feedback over system slowdowns and intrusive automation tools.The company has indicated that upcoming updates will focus on responsiveness, stability and user control, marking a departure from an aggressive push to embed AI across the operating system. The recalibration comes after months of criticism from enterprise customers and everyday users who reported increased resource usage, inconsistent performance and

Duos Edge AI has expanded its footprint in Texas with the launch of a second edge data centre in Amarillo, signalling a push to localise artificial intelligence infrastructure and reduce latency for enterprise users in underserved regions.The new facility, announced this week, is designed to support high-performance computing workloads that require rapid data processing at the network’s edge, including real-time analytics, computer vision and industrial automation. By situating infrastructure closer to end users, the company aims to address persistent bottlenecks

Escalating investments in artificial intelligence infrastructure are driving a fundamental shift in how global technology companies design, build and operate data centres, as demand for computing power accelerates beyond traditional digital services.Major technology firms are committing tens of billions of dollars to expand AI-ready facilities, reflecting a race to secure capacity for training and deploying advanced machine learning models. This surge is not only redefining the economics of cloud computing but also reshaping supply chains, energy consumption patterns and geopolitical

Barcelona-based Qilimanjaro Quantum Tech has introduced a new cloud service designed to broaden access to quantum computing, positioning itself within a fast-evolving race to commercialise next-generation processing power. The platform, branded SpeQtrum QaaS, offers remote access to what the company describes as a multimodal quantum data centre that integrates different quantum technologies under a single interface.The launch signals a shift in how quantum resources are being packaged and delivered, as companies attempt to move beyond experimental systems towards scalable, user-friendly

Crypto. com has cut roughly 12 per cent of its global workforce, targeting positions deemed misaligned with the company’s accelerating integration of artificial intelligence, marking one of the most pronounced restructurings in the digital asset sector as firms recalibrate for automation-led operations.The Singapore-based cryptocurrency platform confirmed that the reductions were part of a broader strategic shift aimed at improving operational efficiency and aligning talent with emerging technological priorities. Roles affected were described internally as those unable to “adapt” to evolving

SoftBank Group is advancing plans for a large-scale artificial intelligence data centre in Ohio, aiming to anchor one of the most energy-intensive computing hubs in the United States with electricity generated predominantly from natural gas, a strategy that underscores the growing tension between AI expansion and climate goals.The Japanese conglomerate is exploring development on federally owned land, with early estimates indicating the facility could require energy infrastructure worth about $33 billion. The project reflects surging global demand for high-performance computing

Meta has unveiled a new initiative offering guaranteed payments and enhanced visibility to content creators, signalling an aggressive push to draw talent away from rival platforms TikTok and YouTube and reinvigorate activity on Facebook.The programme, branded as a “Creator Fast Track”, is designed to attract high-performing influencers by providing upfront financial incentives alongside algorithmic boosts intended to expand their reach. The move reflects intensifying competition among social media companies for creator loyalty, as short-form video continues to dominate user engagement

A malicious Python package masquerading as a legitimate Telegram development tool has been identified as a vehicle for remote code execution attacks, raising concerns about supply chain security within open-source ecosystems.Cybersecurity researchers have flagged a package named “pyronut” that was uploaded to the Python Package Index, presenting itself as a functional alternative to Pyrogram, a widely used Telegram MTProto API framework relied upon by developers to build bots and automated user accounts. Instead of delivering expected functionality, the package embeds

OpenAI has introduced two smaller artificial intelligence models, GPT-5.4 mini and GPT-5.4 nano, signalling a push towards lower-cost, faster systems designed for large-scale deployment and increasingly complex agent-based applications.The announcement marks a shift in emphasis from raw model size to efficiency and adaptability, as enterprises seek to integrate AI into everyday workflows without incurring prohibitive computational costs. The new models are positioned as lighter alternatives within the GPT-5.4 family, optimised for latency-sensitive environments and capable of supporting distributed “subagent” architectures

Cloud infrastructure provider Vultr has introduced a production-ready artificial intelligence inference stack built on NVIDIA’s Rubin platform, marking a significant expansion of their collaboration as enterprises accelerate adoption of generative AI workloads.The deployment is designed to provide scalable, cost-efficient inference capabilities across Vultr’s global cloud network, targeting businesses seeking to operationalise AI models without the high capital expenditure typically associated with dedicated infrastructure. The Rubin platform, positioned as NVIDIA’s next-generation architecture for AI inference, focuses on delivering higher throughput and

ByteDance has halted plans for a global release of its advanced AI video generator, Seedance 2.0, following mounting legal pressure from major film studios concerned about copyright violations and the unauthorised use of creative content.The decision reflects growing tension between technology firms racing to commercialise generative artificial intelligence and entertainment companies seeking to protect intellectual property. Industry executives familiar with the matter indicate that negotiations and internal reviews are under way, with the company reassessing legal exposure before proceeding with

X has started introducing a long-anticipated “dislike” button, marking a notable shift in how users can engage with content on the platform formerly known as Twitter. The feature, which had been tested in limited formats over the past few years, is now being rolled out more broadly, according to updates shared by the company and early user feedback.The move signals a recalibration of X’s interaction design under the leadership of Elon Musk, with the platform seeking to refine how users

Tumblr has rolled back controversial changes to its reblog system following a wave of user protests, marking another instance of the platform reversing course after community pushback.The microblogging service, owned by Automattic, confirmed it would restore key elements of its traditional reblogging functionality after users criticised recent updates that altered how posts are shared and interacted with across the platform. The decision comes amid sustained complaints from long-time users who argued the changes undermined Tumblr’s core identity as a community-driven

TPConnects Technologies has introduced a Model Context Protocol integration within its Astra NDC platform, positioning the Dubai-based firm at the forefront of efforts to standardise how artificial intelligence systems interact with airline retailing infrastructure. The move signals a shift in how carriers and travel sellers deploy AI-driven tools, with the company describing the MCP layer as a step towards more interoperable and context-aware digital commerce in aviation.The integration enables AI agents to communicate with airline retail systems through a

CEOs rarely reach the top by accident. They get there because they make difficult decisions under pressure, synthesize complex information quickly, and move forward when uncertainty would slow others down.Yet the very structure of leadership can quietly undermine those strengths.Research referenced in Harvard Business Review suggests that nearly 60% of CEOs acknowledge that isolation at the top affects their effectiveness, largely because honest feedback becomes less frequent as authority grows. At the same time, global studies from McKinsey show that

Meta Platforms has agreed to spend as much as $27 billion over five years to secure access to high-performance artificial intelligence infrastructure from Nebius Group NV, a move that underscores the escalating race among technology companies to expand computing capacity for advanced AI systems.The agreement gives Meta long-term access to specialised cloud infrastructure built around high-performance graphics processing units designed for training and running large-scale artificial intelligence models. The deal forms part of the company’s broader strategy to strengthen computing

Adobe has agreed to provide free services worth about $75 million to customers as part of a settlement resolving a United States lawsuit over its subscription cancellation practices, closing a dispute that drew scrutiny from regulators and consumer advocates over how the software giant managed its widely used Creative Cloud plans.The agreement follows allegations that Adobe made it unnecessarily difficult for customers to cancel subscriptions, particularly those enrolled in annual plans paid monthly. Authorities argued that users were not adequately

Alibaba Group Holding is preparing to introduce an artificial intelligence service designed to help companies deploy autonomous digital assistants, positioning the technology giant to capitalise on growing demand in China for so-called “agentic AI” systems capable of carrying out real tasks with minimal human input.The planned service will allow corporate users to build and manage AI agents that can automate business processes such as data analysis, scheduling, workflow coordination and digital customer support. The move reflects intensifying competition among

JD. com has launched a sweeping push into Europe, opening its online retail platform across several major markets including the United Kingdom, marking a strategic attempt by the Chinese e-commerce giant to compete with Amazon and diversify beyond its highly competitive domestic market.The company introduced its international marketplace under the Joybuy brand in the UK, Germany, France, the Netherlands, Belgium and Luxembourg, offering a wide range of products such as electronics, home appliances, beauty items, groceries and household goods. The

OpenAI is preparing to expand the capabilities of its widely used chatbot by bringing advanced video-generation technology from its Sora model into ChatGPT, a move that could transform how users interact with artificial intelligence and reshape the competition among global AI platforms.Plans under discussion within the company involve integrating Sora’s text-to-video generation system directly into ChatGPT, allowing users to create short videos simply by typing prompts into the chatbot interface. The initiative aims to broaden the platform’s multimedia abilities and

Google has finalised its $32 billion acquisition of the cloud security firm Wiz, marking one of the largest deals in the cybersecurity sector and underscoring the growing importance of protecting complex cloud infrastructure as companies expand their use of artificial intelligence and distributed computing.The purchase brings Wiz into the Google Cloud division while allowing the company’s technology to continue operating across multiple cloud platforms. Executives from both firms have said the strategy reflects a shift among enterprise customers that increasingly

Cybersecurity researchers have demonstrated a method to circumvent safety guardrails embedded in widely used generative artificial intelligence systems, raising concerns about the reliability of protective controls designed to prevent misuse of large language models.A research team from Palo Alto Networks’ Unit 42 disclosed that a specially crafted attack can bypass safeguards in some generative AI platforms by manipulating how the models interpret safety instructions. The technique, known as “Bad Likert Judge,” prompts the AI system to evaluate harmful content on

Artificial intelligence is set to reshape the software industry, yet the transformation is unlikely to lead to a sweeping decline in credit ratings across the sector, according to an assessment by S&P Global Ratings that seeks to temper investor concerns about widespread financial stress among software companies.Analysts at the credit ratings agency say the rapid development of AI technologies may alter competitive dynamics and business models for many software firms, but the credit impact is expected to vary widely between

Autonomous delivery robots are beginning to move beyond pavements and into urban streets, as robotics company Coco Robotics deploys a faster generation of machines guided by mapping technology originally developed for the global mobile game Pokémon GO.The company says its latest machine, known as Coco 2, can travel at speeds of up to 13 miles per hour while navigating city roads and bicycle lanes where local regulations permit. The system relies on spatial mapping and visual positioning tools from Niantic

Heightened geopolitical tension across the Middle East has triggered a wave of digital scams and coordinated misinformation campaigns, prompting cybersecurity specialists to warn that conflict-related narratives are being exploited to deceive the public through emails, messaging services and social media platforms.Security analysts report a sharp increase in phishing attempts, fake fundraising drives and manipulated online content that claim links to unfolding events across the region. The campaigns, which often mimic legitimate humanitarian appeals or urgent security updates, are designed to

[caption id="attachment_114114" align="aligncenter" width="781"] Anthropic Logo[/caption]Legal confrontation between artificial intelligence developer Anthropic and the United States Department of Defense has triggered an unusual show of solidarity across the technology sector, with researchers from competing companies publicly supporting the firm’s challenge against a national security designation that threatens its business.Anthropic filed lawsuits in federal courts against the Pentagon and other government agencies after officials classified the company as a “supply-chain risk”, a label that effectively prevents defence contractors from working with

Chinese authorities have begun restricting the use of OpenClaw artificial-intelligence applications on office computers within banks, government departments and state-owned enterprises, responding to mounting security concerns surrounding the emerging technology.Instructions circulated through regulatory and internal administrative channels direct public institutions and state-controlled companies to halt installation or use of OpenClaw AI tools on official devices and internal networks. The move reflects a tightening approach to generative AI platforms that operate outside the country’s tightly regulated technology ecosystem.Officials overseeing financial institutions

Chinese artificial intelligence developer DeepSeek has withheld early access to its newest large language model from two of the world’s dominant semiconductor designers, Nvidia and AMD, a move that industry observers interpret as a calculated step in Beijing’s effort to reduce dependence on United States technology and strengthen domestic computing ecosystems.The decision centres on the company’s latest generation AI model, which is designed to compete with leading systems developed by firms in the United States. Engineers at Nvidia and AMD,