Topic
Explore English articles about AI Agents.
The Linux Foundation has begun foundation operations for "x402," a common specification that allows AI to make payments directly via HTTP. With the protocol transferred from Coinbase, 40 organizations—including major card companies and cloud providers—have joined, aiming to standardize and popularize machine-to-machine payments.
Customers worried about tight supply and rising prices for memory and other components amid AI demand redirected their capital spending toward servers and storage, delaying IBM's mainframe-related deal closures. This caused revenue to fall short of market expectations, sending the company's stock down by its worst-ever margin.
Databricks' internal benchmark revealed a cost reversal: Sonnet 5, despite its lower per-token price, cost more per task than Opus 4.8. We unpack the structure—driven by completion rates and harness selection—alongside Uber's budget overrun case.
SK hynix's CEO has forecast the worst memory supply shortage in history for 2027. While AI demand spreads beyond HBM into DRAM and NAND, the expanded production facilities in Yongin and Cheongju will only come online gradually between 2027 and 2029.
OpenAI announced that roughly 30% of the AI coding benchmark "SWE-Bench Pro" contains flaws, withdrawing its recommendation. Because defects such as failing correct solutions prevent accurate model evaluation, continuous auditing of the grading system itself—before comparing rankings—becomes essential going forward. (123 characters)
Meta has released an API for its flagship AI model "Muse Spark 1.1," which significantly strengthens coding and long-running agent performance. By shifting to a strategy of keeping model weights private and setting usage-based pricing far below competitors, Meta aims to capture the developer market.
OpenAI has released the new model GPT-5.6, rolling out three tiers with different capabilities and price points. Rather than a simple price cut, it introduces mechanisms for controlling reasoning volume and parallel agent execution, shifting the axis for evaluating AI from per-token pricing to the total cost of completing a job.
Microsoft has released TypeScript 7.0, reimplemented in Go, delivering roughly 8x to 12x faster type checking and other processing compared to previous versions. Optimized parallel processing boosts efficiency for large-scale development, while compatibility with existing APIs means the older version is recommended for continued side-by-side use.
To control tritium formation in molten salt—essential for fusion reactors to breed their own fuel—a new method combining quantum computers with classical computation has been developed. By simulating the complex behavior of electrons with high precision, it tackles the prediction of corrosive compound formation and efficient fuel recovery.
Entire, founded by former GitHub CEO Thomas Dohmke, has unveiled a preview of its decentralized Git network. As massive access from AI agents exposes GitHub's limits, the verifiability of the performance numbers and the involvement of Microsoft-affiliated investors take center stage.
