Generative models, prompting, agents and AI tool capabilities.
Do Claude and GPT Get Nerfed After Launch? We Have Proof Now
26d ago · English audio
One week after Opus 5.5 launched, a Reddit thread with 2,691 upvotes said Anthropic had nerfed it. This time a daily benchmark was keeping score. This video shows what the nerf trackers saw and how to build one of your own. Anthropic and OpenAI both say the weights behind a pinned model ID never change. The router, the safety classifiers, the prompts and the harness around the model change all the time. You'll see…
Devsplainers
AI tools and the tech behind them, explained in plain English. ChatGPT, Claude, Gemini, AI that runs on your own computer, and the AI coding tools developers use every day. Which tool should you use, what it really costs, whether the paid plan is worth it, or whether the free option is good enough. We'll give you the verdict you need for your work. You don't need to be a senior engineer to follow along. If you use…
Outlier score
This score was measured on Sat, 31 Oct 2026 07:11:23 GMT. Current view counts may differ.
57.1K views ÷ a 16.7K baseline = 3.4×. Compared with 50 earlier videos from the same channel, matching the format recorded at scoring time. Older-upload baseline: reference videos had at least as long to gain views. This comparison is not age-matched.
YouTube snapshot: Sat, 31 Oct 2026 10:54:30 GMT. Counters may have changed.
Read the methodologyGenerative AI & Agents
Channel-balanced percentile: 86
Comparison details
97 videos from 72 other channels, with each channel weighted equally. Same format and language, with 10,000–99,999 subscribers. Observed at 628 hours after publication, within ±48 hours.
This describes the collected sample, not all of YouTube. Updated Sat, 31 Oct 2026 11:33:07 GMT.
Niches & tags
AI evaluation, societal implications, governance and responsible deployment.
Transcripts and detailed analytics are available for recent outlier Shorts up to three minutes. Research tools