The latest technology news from across Asia-Pacific
The EU's government satcom program went operational January 14, pooling capacity from existing national GEO satellites across five member states. It's a bridge to IRIS², the 290-satellite constellation now targeting 2029 instead of 2030. The real test: whether shared capacity holds up during actual crises.
Security researchers documented 230 malicious OpenClaw "skills" disguised as crypto trading tools uploaded to ClawHub since late January. The extensions exploit OpenClaw's unsandboxed architecture to steal browser data and cryptocurrency information. Enterprise deployments are exposed: hundreds of instances run online without authentication.
A consortium led by KKR is in advanced talks to acquire a majority stake in ST Telemedia Global Data Centres, valuing the Singapore operator at over US$10 billion including debt. The deal, which would include Singtel, GIC, and Mubadala as minority investors, reflects the AI-driven surge in Asia-Pacific data center demand.
A former Google employee alleges the company violated its own AI ethics policies by helping an Israeli defense contractor improve object recognition in drone footage. The complaint claims Google misled investors, while the company argues the support was too limited to matter. Google revised its AI ethics policy months after the incident.
Apple is considering a flip-style foldable iPhone to follow its first fold device, while internal questions mount about whether the company has what it takes to compete in AI. New MacBook Pros with M5 Pro and M5 Max chips are set to ship alongside macOS 26.3.
Alibaba is spending 3 billion yuan on Qwen AI chatbot promotions during China's Lunar New Year holiday, outspending Tencent and Baidu combined. The spending blitz - focused on dining vouchers, entertainment, and digital red envelopes - signals how seriously Chinese tech giants are taking AI consumer adoption, even as ROI remains unproven.
Senior Apple leaders are debating whether the company has the right ingredients to win in AI, according to sources. The introspection comes as new M5 Pro and M5 Max MacBook Pros prepare to ship and Apple explores a clamshell foldable iPhone to follow its first fold device.
YC-backed Clearspace is recruiting an ML researcher ($150K-$200K, 0.5-1% equity) to build production models that filter network traffic based on natural language rules. The hire signals a shift from ScreenTime API interventions to deeper behavioral classification—a technical bet worth watching as digital wellness tools mature.
A community wiki preserves Commodore's 1990 AT&T Unix port for Amiga hardware - a commercial failure that shipped around 1,000 beta units before Commodore's 1994 collapse. The project highlights early challenges porting Unix to consumer hardware, now emulatable via WinUAE since 2013.
Google Research tested 180 agent configurations across five architectures and found multi-agent systems boost parallel tasks by 81% but degrade sequential work by up to 70%. The predictive model correctly identifies optimal setups for 87% of unseen tasks—challenging the 'more agents always better' assumption driving enterprise AI deployments.
The New York Times announced Wordle will start repeating previous answers from February 2, framing it as 'exciting news.' Independent analysis suggests the original 2,309-word list could have lasted until 2036 with proper curation. The move fundamentally changes player strategy and has drawn skepticism from the community.
Europe's largest open-source conference shifted focus from vendor convenience to community-controlled infrastructure. The 25-year-old event showcased 1,079 sessions across 71 tracks, with notable emphasis on FreeBSD, minimal BSD systems, and decentralized networking—signaling enterprise interest in sovereign tech stacks.