Search “Anthropic”

Anthropic proposes using value axes to compress and study Claude's values

On July 13, 2026, Anthropic proposed a method for studying Claude's values by compressing them along value axes: using four value axes—compliant vs. cautious, warm vs. rigorous, depth vs. concise, and candid vs. execution—to compress the thousands of values expressed by Claude, with these four axes explaining 15% of the variance in values. The method is also used to compare different models and languages: Opus 4.6 leans toward compliant, rigorous, concise, and execution, while Opus 4.7 leans toward cautious, rigorous, depth, and candid; Sonnet 4.6 is warmer, consistent with users' subjective impressions. The largest cross-language difference is on the warm vs. rigorous axis, with Arabic and Hindi leaning most toward warm, and English and Russian leaning most toward rigorous. There are currently no follow-up developments.

View story:Anthropic proposes using value axes to compress research on Claude's values

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

English translation is being prepared.

View story:English translation is being prepared.

English translation is being prepared.

View story:English translation is being prepared.

Anthropic tests language model performance on robotic tasks

Anthropic released research on July 9, 2026, testing the performance of multiple language models on classic control, locomotion navigation, and robotic arm manipulation tasks, and built the Embody benchmark suite. The research results show that model capability is highly dependent on the connection method with the robot: when directly driving joints, most fail, while when using a supervised pretrained controller or a simple orientation tool, the models can complete real navigation and manipulation tasks. In LIBERO direct manipulation, the full task success rate is only 0 to 5.5%; after connecting to a VLA policy, the success rate of each model increases substantially, but the strongest model still falls short of MolmoAct running alone. There is currently no follow-up progress.

Anthropic tests language model performance on robotic tasks

Anthropic released research on July 9, 2026, testing the performance of multiple language models on classic control, locomotion navigation, and robotic arm manipulation tasks, and built the Embody benchmark suite. The research results show that model capability is highly dependent on the connection method with the robot: when directly driving joints, most fail, while when using a supervised pretrained controller or a simple orientation tool, the models can complete real navigation and manipulation tasks. In LIBERO direct manipulation, the full task success rate is only 0 to 5.5%; after connecting to a VLA policy, the success rate of each model increases substantially, but the strongest model still falls short of MolmoAct running alone. There is currently no follow-up progress.

View story:Anthropic tests language model performance on robotic tasks

English translation is being prepared.

English translation is being prepared.

View story:English translation is being prepared.

English translation is being prepared.

View story:English translation is being prepared.

Anthropic adjusts the Economic Index data pipeline

Anthropic adjusted the data pipeline for its Economic Index, switching to privacy-preserving telemetry data collected continuously on a daily basis, thereby making it possible to observe hourly usage patterns. This change was released together with the sixth issue of the Economic Index report. The report shows that the share of conversations for personal use rose from about 35% on weekdays to nearly 50% on weekends; recipe requests show a 2.3-fold peak at 18:00; and tax-related conversations on April 14 were 8 times those on an average day in May. The report also classifies conversations by artifact, with artifacts identified in 93% of conversations, of which explanatory ones account for 17% and documents and reports account for 15%. At present, this event has only the above first-party report as evidence, and there have been no subsequent developments.

Anthropic announces the results of the second phase of the Project Fetch experiment

On June 18, 2026, Anthropic announced the results of the second phase of the Project Fetch experiment. The experiment showed that Claude Opus 4.7, running Claude Code without human assistance, completed tasks that human teams had done last year about 20 times faster than the fastest human team. On every task that at least one human team completed, Opus 4.7 was at least 10 times faster; on tasks completed by all four human teams, it was on average more than 37 times faster than teams without Claude and more than 18 times faster than teams with Claude, while generating nearly ten times less code than Team Claude. Anthropic also noted that the model still fails at the closed-loop control step of precisely pushing the beach ball, that this is not a solved robotics problem, and that the improvements came from general capability scaling rather than robot-specific training.

Anthropic adjusts Economic Index data pipeline

Anthropic adjusted the data pipeline for its Economic Index, switching to privacy-preserving telemetry data collected continuously on a daily basis, thereby making it possible to observe hourly usage patterns. This change was released together with the sixth issue of the Economic Index report. The report shows that the share of conversations for personal use rose from about 35% on weekdays to nearly 50% on weekends; recipe requests show a 2.3-fold peak at 18:00; and tax-related conversations on April 14 were 8 times those on an average day in May. The report also classifies conversations by artifact, with artifacts identified in 93% of conversations, of which explanatory ones account for 17% and documents and reports account for 15%. At present, this event has only the above first-party report as evidence, and there have been no subsequent developments.

View story:Anthropic adjusts the Economic Index data pipeline

English translation is being prepared.

View story:English translation is being prepared.

Anthropic announces results of Project Fetch Phase 2 experiment

On June 18, 2026, Anthropic announced the results of the second phase of the Project Fetch experiment. The experiment showed that Claude Opus 4.7, running Claude Code without human assistance, completed tasks that human teams had done last year about 20 times faster than the fastest human team. On every task that at least one human team completed, Opus 4.7 was at least 10 times faster; on tasks completed by all four human teams, it was on average more than 37 times faster than teams without Claude and more than 18 times faster than teams with Claude, while generating nearly ten times less code than Team Claude. Anthropic also noted that the model still fails at the closed-loop control step of precisely pushing the beach ball, that this is not a solved robotics problem, and that the improvements came from general capability scaling rather than robot-specific training.

View story:Anthropic announces the results of the second phase of the Project Fetch experiment

English translation is being prepared.

View story:English translation is being prepared.

English translation is being prepared.

View story:English translation is being prepared.