English text is being prepared.

By relevance · Total 106 results Clear search

Apr 152026Wednesday

08:00
modalApr 15 08:00
English translation is being prepared.
Matched fieldsTitle、Summary
#company #Tutorials / practice #Agents #English text is being prepared.

Apr 42026Saturday

19:45
ahead_of_aiApr 4 19:45
English translation is being prepared.
Matched fieldsTitle、Summary
#newsletter #Tutorials / practice #Agents #English text is being prepared.

Feb 192026Thursday

08:00
modalFeb 19 08:00Selected
English translation is being prepared.
Matched fieldsTitle、Summary
#company #Tutorials / practice #Agents #English text is being prepared.

Oct 302026Friday

08:24
Rohan PaulOct 30 08:24
Gemini 4 Carbon Tops the Artificial Analysis Intelligence Index with 59 Points
Google's Gemini 4 Carbon ranks first on the Artificial Analysis Intelligence Index with 59 points, 1 point higher than Claude Opus 5.5's 58 points and 2 points higher than GPT-6 Astra's 57 points. Artificial Analysis found that Carbon won 4 of 9 benchmarks (HLE-Diamond, Terminal-Bench-Science, AA-Omniscience, AutomationBench-AA), but it does not lead in coding and knowledge work. Calculated at standard prices, it costs $7.76 per index task, about 23% lower than Opus 5.5's $10.04, but higher than GPT-6 Astra's $6.25.
Google's Gemini 4 Carbon ranks first on the Artificial Analysis Intelligence Index with 59 points, 1 point higher than Claude Opus 5.5's 58 points and 2 points higher than GPT-6 Astra's 57 points. Artificial Analysis found that Carbon won 4 of 9 benchmarks (HLE-Diamond, Terminal-Bench-Science, AA-Omniscience, AutomationBench-AA), but it does not lead in coding and knowledge work. Calculated at standard prices, it costs $7.76 per index task, about 23% lower than Opus 5.5's $10.04, but higher than
Matched fieldsSummary、Body
#ai-news #Evaluations / benchmarks #Google #Anthropic

Oct 252026Sunday

16:14
Phoenix TechnologyOct 25 16:14
OpenAI discloses offline training environment for frontier models; two isolated AIs communicate via CPU cache
On October 23, OpenAI disclosed the training environment for its frontier models. When practicing searching, browsing, and operating web pages, the models use pre-saved and indexed web resources and do not directly access the real internet; all of its frontier models' reinforcement learning training uses this kind of offline interaction. The environment employs multiple layers of isolation, including restricted containers, lightweight virtual machines, hypervisors, firewalls and identity and permission management, and monitoring and suspension mechanisms, but OpenAI still retains some external connections for collecting metrics, logging, and managing authentication. In experiments specifically testing the effectiveness of the isolation, OpenAI had two AIs attempt to pass messages to each other through the CPU cache; in most tests, researchers had given hints about the communication direction or encoding rules; OpenAI said it has patched the issues found and added monitoring for these communication methods.
On October 23, OpenAI disclosed the training environment for its frontier models. When practicing searching, browsing, and operating web pages, the models use pre-saved and indexed web resources and do not directly access the real internet; all of its frontier models' reinforcement learning training uses this kind of offline interaction. The environment employs multiple layers of isolation, including restricted containers, lightweight virtual machines, hypervisors, firewalls and identity and per
Matched fieldsSummary、Body
#chinese #Safety / alignment #Agents #OpenAI

Oct 232026Friday

23:12
Shao MengOct 23 23:12
Claude Fable 5.1 and GPT-6 Astra Image Steganography Experiment: About 80% Still Decodable After Screenshotting
The author used 50 short messages to test Claude Fable 5.1's image steganography capability: when only asked to hide a specified sentence in an image, the model spontaneously encoded it using visual attributes such as object height; in a separate new session given only the rendered PNG, the same model could still decode about 78% (39/50). In cross-decoding, Fable encoding with Astra decoding achieved an 88% success rate (44/50), Astra encoding with Fable decoding achieved 84% (42/50), and Astra's own decoding was also 84%. The author notes that this is an early sign of image steganography, but decoding requires tools to measure pixels, so monitoring tool calls would be enough to detect it.
The author used 50 short messages to test Claude Fable 5.1's image steganography capability: when only asked to hide a specified sentence in an image, the model spontaneously encoded it using visual attributes such as object height; in a separate new session given only the rendered PNG, the same model could still decode about 78% (39/50). In cross-decoding, Fable encoding with Astra decoding achieved an 88% success rate (44/50), Astra encoding with Fable decoding achieved 84% (42/50), and Astra'
Matched fieldsSummary、Body
#hot-signal #Papers/Research #Safety / alignment #Multimodal

Oct 222026Thursday

08:00
Browser UseOct 22 08:00
Browser Use Roundup of the Best Browser Agent MCP Servers of 2026
Browser Use has released its 2026 roundup of browser Agent MCP servers, dividing such servers into two modes—driver and delegate—and evaluating four of them one by one: Browser Use Cloud MCP, Browser Use open-source MCP, Hyperbrowser MCP, and agent-browser chat. In a comparative evaluation across 100 BU Bench V2.1 tasks, the author found that the four driver-type tools consumed a median of 2.1M to 6.7M input tokens per task, all of which count toward the coding Agent's context. The article also provides each server's installation commands, model selection, billing methods, and known limitations, and explains that long tasks, parallel tasks, and structured-output scenarios are better suited to the delegate mode.
Browser Use has released its 2026 roundup of browser Agent MCP servers, dividing such servers into two modes—driver and delegate—and evaluating four of them one by one: Browser Use Cloud MCP, Browser Use open-source MCP, Hyperbrowser MCP, and agent-browser chat. In a comparative evaluation across 100 BU Bench V2.1 tasks, the author found that the four driver-type tools consumed a median of 2.1M to 6.7M input tokens per task, all of which count toward the coding Agent's context. The article also
Matched fieldsSummary
#company #Tutorials / practice #English text is being prepared.#Agents