Wiredframe Radar logo

Category intelligence

Social Media Briefing — July 25, 2026

9 current items analyzed and ranked.

Executive synthesis

Social Media Summary

Model releases and frontier evaluation insights led discussions today. Sakana AI announced Fugu-Ultra v1.1, using dynamic orchestration to tackle complex reasoning tasks, while hands-on testing of Opus 5 generated significant community interest.

Key Themes

Model Releases & Updates · 3AI Policy & Geopolitics · 1AI Evaluation & Research · 3

Primary evidence

Top Ranked Signals

Social Bluesky Jul 24

Our team just shipped Fugu-Ultra v1.1! 🐡 By dynamically orchestrating the latest frontier models, w...

By @hardmaru.bsky.social

85 score
AI Analysis

Continuing our coverage from yesterday, David Ha announces the release of Fugu-Ultra v1.1 by Sakana AI, noting it outperforms Fable 5 in complex reasoning through dynamic model orchestration.

Our team just shipped Fugu-Ultra v1.1! 🐡 By dynamically orchestrating the latest frontier models, we pushed performance up by 7.9 points. We are now beating Fable 5 in complex coding and reasoning tasks without even having Fable 5 in our agent pool. Collective intelligence is the future.
Model Releases & UpdatesCollective Intelligence
Social Bluesky Jul 24

Announcing Fugu-Ultra v1.1 🐡 We’ve been thrilled by the reception to the Fugu model family. Thanks ...

By @sakanaai.bsky.social

82 score
AI Analysis

Continuing our coverage from yesterday, Sakana AI officially announces Fugu-Ultra v1.1, incorporating the latest frontier models based on user feedback.

Announcing Fugu-Ultra v1.1 🐡 We’ve been thrilled by the reception to the Fugu model family. Thanks to everyone who tried it, shared feedback, and trusted Fugu with real work. Today, we’re releasing Fugu-Ultra v1.1 → sakana.ai/fugu Upgraded to incorporate the latest frontier models.
Model Releases & Updates
Social Mastodon (dair-community.social) Jul 24

Its all about "free markets" and "let the market decide" until you have competition that shows how f...

By @timnitGebru@dair-community.social

80 score
AI Analysis

Timnit Gebru critiques OpenAI and Anthropic for lobbying government intervention against incoming Chinese AI model competition.

Its all about "free markets" and "let the market decide" until you have competition that shows how full of bullshit you are and then you run to daddy Trump saying to stop these bad Chinese models from entering our market.The irony of OpenAI and Anthropic running to the government to save them whenever there’s competition and it’s this government that complains about “handouts” and communism.
AI Policy & Geopolitics
Social Bluesky Jul 24

Unexpected finding from a study on ChatGPT's impact on college: "once the COVID-19 disruption is mod...

By @emollick.bsky.social

75 score
AI Analysis

Ethan Mollick highlights an academic study finding that ChatGPT's introduction had no detectable effect on college grades once COVID-19 disruptions are controlled.

Unexpected finding from a study on ChatGPT's impact on college: "once the COVID-19 disruption is modeled separately, the introduction of ChatGPT had no detectable effect on grades.. course evaluations for subject understanding, interest, and relative workload show no change" arxiv.org/pdf/2607.21534
AI in Education & Research
Social Bluesky Jul 24

As a joke I prompted Codex "Build and run BenchBench, a benchmark of now good ai is at creating benc...

By @emollick.bsky.social

72 score
AI Analysis

Ethan Mollick experiments with Codex to create 'BenchBench' (a benchmark for AI benchmark generation) and notes the unexpected quality of the resulting paper.

As a joke I prompted Codex "Build and run BenchBench, a benchmark of now good ai is at creating benchmarks. then figure out what benchbenchbench is and run that. and then write benchbenchbench up as a good arXiv paper." I got a PDF. But the paper is actually kind of interesting? Weird.
AI BenchmarkingAgentic Workflows
Social Bluesky Jul 24

As far as I can tell their sandbox was a container with network access denied except for a single IP...

By @simonwillison.net

70 score
AI Analysis

Simon Willison analyzes a sandbox container network proxy flaw and references his previous writing on the topic.

As far as I can tell their sandbox was a container with network access denied except for a single IP for an HTTP proxy that controlled which sites could be accessed, and the proxy turned out to have a flaw I wrote about their production version of that back in Jan: simonwillison.net/2026/Jan/26/...
AI Security & Sandboxing
Social Bluesky Jul 24

Glad to see Google sharing data on how Gemini is being used. Especially interesting is that the usef...

By @emollick.bsky.social

68 score
AI Analysis

Ethan Mollick discusses Google's data showing multimodal AI's unexpected usefulness in manual labor contexts.

Glad to see Google sharing data on how Gemini is being used. Especially interesting is that the usefulness of multimodal AI for manual labor may be greater than expected. blog.google/innovation-a...
Multimodal AIReal-World Applications