Inherent, founded by DeepMind alumni, says its AI ‘teammate’ just outperformed Anthropic and OpenAI at replicating research
THE SO WHAT
If Faraday can reliably replicate scientific papers, the bottleneck in R&D shifts from reading the literature to designing the next experiment. For teams in pharma, materials, or frontier research, the question this week is whether your stack is ready to plug into agents that treat the entire paper ecosystem as an executable environment, not a PDF archive.
READ THE SOURCE
MORE FROM THE WIRE
Applied AINvidia Customers Notified About AI-Related Price Hikes Above 15%
A 15%+ jump in AI server pricing with memory as the pressure point means your unit economics for model training and high-throughput inference just got worse overnight. Operators should re-run ROI on planned GPU-heavy projects this week and accelerate work on efficiency levers — pruning, quantization, SLMs, and better scheduling — before committing to new capacity.
Applied AIBefore you go back to school, set up these 5 ChatGPT tools — your future self will thank you
Students normalizing ChatGPT for planning, comprehension, and study workflows is a preview of how the next cohort will expect to work inside your org. If you hire early-career talent, assume AI-native habits and focus your enablement on guardrails, verification, and data hygiene rather than basic tool training.
Applied AIClaude wins out in major user satisfaction survey, beating Gemini and ChatGPT — but it's bad news for Grok and Siri
User satisfaction skewing toward Claude over Gemini and ChatGPT is a reminder that perceived reliability and UX — not just raw benchmarks — are becoming a real moat. If you’re standardizing on a primary assistant, measure internal NPS and task completion, not just model scores, before you lock in contracts.
Applied AIApollo finds AI is hitting paychecks rather than payrolls
AI exposure is quietly repricing labor before it destroys it—wage growth in highly exposed roles slowed 6.7% post‑2023, with a 10.7% drag in the lowest‑paid quartile and no hit at the top. If you run a team in that exposed middle, expect retention pressure and morale risk long before headcount reductions show up in your plan.