Claude Opus 4.8 is Here — Should You Switch?
💼 Business owner or operator with a team? We build AI automation systems that cut costs and scale ops — done for you: https://ryanandmattdatascience.com/ai-consultant/ 🚀 Want to make money with AI skills? Join our free community — real projects, real client strategies, and the exact stack we use: https://www.skool.com/data-and-ai Anthropic just released Claude Opus 4.8. Today we are going to look at the recent article and system card that was released. How does Claude Opus 4.8 compare to other models like Sonnet, GPT 5.6 and others. Is Claude Mythos being released soon? TIMESTAMPS 00:00 - Intro: Claude Opus 4.8 just dropped 00:33 - Anthropic's announcement post & pricing (same as 4.7) 01:00 - New feature: effort control (fast/normal modes) 01:30 - Dynamic features & 3x cheaper claims 01:50 - Benchmark comparison vs GPT-5.5 & OpenAI 02:25 - Agentic terminal performance 03:00 - Coding & SWE-bench improvements 04:00 - What testers are saying 05:15 - Release details & scoring methodology 06:30 - Subagent rollout (first big new feature) 07:25 - Spinning out subagents & planning 08:00 - Cloud Code integration & mid-task instructions 09:20 - Default sub-agent depth settings 10:00 - Availability across platforms 10:30 - Diving into the model card 11:10 - Where Claude 4.8 still falls short 11:30 - Failure examples walkthrough begins 12:30 - False "babysitting" / phantom requests issue 13:00 - Wrong API & missing context failures 14:00 - Repeated failed approaches 15:10 - Billing test failure example 16:00 - List/access investigation failures 17:15 - Transcript viewer / README task failures 18:00 - Investigation based on assumptions 19:00 - Automated AI R&D capability evaluation 20:00 - Prompt injection results 20:40 - Comparison to Gemini models 21:10 - Why this matters (75% blocked) 22:00 - Thinking models & injection vulnerability 23:00 - 4.8 actually worse than 4.7 at prompt injections 23:50 - Informativeness regression 24:25 - Final thoughts 24:55 - Skool group plug & outro OTHER SOCIALS: 🌐 Website & Blog: https://ryanandmattdatascience.com/ Ryan’s LinkedIn: https://www.linkedin.com/in/ryan-p-nolan/ Matt’s LinkedIn: https://www.linkedin.com/in/matt-payne-ceo/ Twitter/X: https://x.com/RyanMattDS *This is an affiliate program. We receive a small portion of the final sale at no extra cost to you.