{{first_name | Leader}}, welcome back.
Last week, Google’s Gemini 3 pulled ahead on key benchmarks and OpenAI reportedly called a “code red.” It set a new tone for how fast the stack is moving, and today’s updates build right on top of that momentum.
Today’s updates:
Save $360 with a full year of Framer Pro, free for early-stage startups.*
AWS launches Nova 2 models, Trainium3 chips, and Frontier AI agents
Nvidia claims 10x AI performance with new multi‑GPU servers
Leading AI companies fall short of global safety standards
Tools, resources, and a prompt for a quick, AI-driven read on how healthy your current pipeline is ⬇️
First impressions matter. With Framer, early-stage founders can launch a beautiful, production-ready site in hours. No dev team, no hassle.
Join hundreds of YC-backed startups that launched here and never looked back.
Key Value Props:
One year free: Save $360 with a full year of Framer Pro, free for early-stage startups.
No code, no delays: Launch a polished site in hours, not weeks, without hiring developers.
Built to grow: Scale your site from MVP to full product with CMS, analytics, and AI localization.
Join YC-backed founders: Hundreds of top startups are already building on Framer.
Eligibility:
Pre-seed and seed-stage startups, new to Framer.
AWS pushes deeper into custom models and agentic workflows
AWS rolled out Nova 2 and Nova Forge, giving teams a clearer path to build their own models instead of relying only on foundation ones. Frontier Agents and Nova Act plug straight into engineering workflows, helping with UI automation, security reviews, and DevOps cleanup. Trainium3 UltraServers are now live too, making large-scale training cheaper and easier to run inside EC2.
For enterprise teams, this is a direct signal: custom models, faster agents, and lower-cost compute are becoming the default, not the edge case. It’s now much easier to go from an idea to a production workflow without overextending your infra or engineering bandwidth.
Nvidia highlights major gains for MoE-heavy workloads
Nvidia shared new benchmarks showing its latest 72-GPU server delivering close to a tenfold performance lift for mixture-of-experts models. These systems are optimized for the routing patterns MoE models need, which means cheaper, faster inference at scale.
If your AI roadmap leans on agents, retrieval, or multimodal systems, this performance jump will influence 2025 infra planning. MoE is becoming mainstream across leading labs, and Nvidia is shaping its hardware around that shift.
New AI safety index shows big gaps across major labs
A new AI Safety Index reviewed leading labs and found all of them falling short of emerging global standards. The issues span transparency, evaluation depth, and readiness for controlling higher-risk systems.
Governance can’t be an afterthought. Internal review protocols, audits, and safety workflows need to mature in parallel with model adoption. It’s not a crisis headline, but it’s a useful direction for how boards and AI teams should plan for long-term responsible use.
Productivity Tools
📊 Optimove – Customer-led platform that personalizes marketing and strengthens engagement across automated journeys
📐 Desmos – Free online math platform offering interactive activities that support teaching and learning
🌐 WebWave – AI website builder enabling users to create polished, professional sites quickly and effectively
📇 Folk – Lightweight AI CRM helping teams organize relationships and streamline workflows with customizable
Get featured tomorrow: How do you use AI for business/personally? Interesting stories will be shared with 100K curious readers.
Useful Resources
How to establish credible thought leadership on YouTube through structure, storytelling, and consistency.
How AI agents are now catching other AI agents hallucinating, improving reliability across systems.
Why “vibe coding” fails in real B2B apps and what to learn to ship production-ready software.
Practical strategies for managing multiple projects without losing clarity or momentum.
How to use Claude Projects to plan and develop higher-quality long-form content.
💰 Funding
Eon raised $300M to scale its AI infrastructure and grow adoption across global enterprises.
Tutor Intelligence raised $34M to expand its AI-driven industrial automation platform.
💼 Roles in AI
🐦 Important differentiator
Pipeline Health Predictor
When to use this?
When you want a quick, AI-driven read on how healthy your current pipeline is — and what actions could help hit next quarter’s revenue target.
You are a revenue operations analyst.
Review the pipeline summary below and assess overall health.
Include:
Win rate projection based on deal mix and stage distribution.
3 early risk signals that could derail target attainment.
3 actions to improve close rates this quarter.
A 1-line summary I can use in leadership updates.
Keep it concise and output in a clean table + short summary paragraph.Correct Input Style:
Pipeline Summary:
Total Pipeline – $48M
Target – $40M
Deals by Stage – 40% early, 35% mid, 25% late
Avg Deal Size – $180K
Win Rate Last Q – 28%
Key Risks – slower enterprise approvals, longer security reviews, hiring freeze.
P.S. Get more such prompts in the Prompting Playbook (free for you)
Q. As leaders, is your board more concerned about AI risk or AI competitiveness?
Stay curious, {{first_name | leaders}}
PS. If you missed yesterday’s issue, you can find it here.