Microsoft just integrated Copilot in OneDrive.
This integration will revolutionize the way we work.
Here's how this new feature will impact your life: ⤵️
Sharing how I think about the AI investing ecosystem:
1. The Application Layer
Integration of AI models into applications like chatbots and assistants.
VCs have interesting opportunities to invest, but little opportunity for public investors.
2. The Model Layer
Three primary developers for AI algorithms: big tech firms, private research firms, and open-source organizations.
The most well-known firms are OpenAI and Anthropic, which big tech has invested heavily in.
Again, not a great avenue for public investors, although they can get exposure through big tech companies.
3. The Data Layer
Long-term, AI is additive to the data space.
Databricks and Snowflake are two most high-profile data providers, providing tools for management of big data.
$AMZN $MSFT $GOOG offer the widest array of products across the data landscape.
4. The Compute Layer
AI compute can be broken down into training and inference.
Training is done in datacenters, mostly using $NVDA chips.
Inference is split between data centers and on-device (mobile and auto for example).
5. The Hardware Layer
Semiconductors dominate the hardware layer.
The semiconductor value chain falls into this segment.
Data center hardware such as servers, networking, and storage is included as well.
6. Investing opportunities
VCs have the easiest access to invest from the top-down.
Public investors have most of the same opportunities to invest from the bottom-up.
My goal as a public investor isn’t to find the next “AI 100 bagger”, it’s to find valuable opportunities with long-term exposure to AI.
“Too often, when we comment on outputs, we are simply stating our uninformed opinion or personal preference.” - @ttorres
https://t.co/qF2x6P9ekD
#prodmgmt#ux#engineering
“Showing your work helps you manage stakeholders throughout the discovery process. These same tactics will help you stay aligned with your team while you are working from home.” - @ttorres
https://t.co/kz761IT8jn
#prodmgmt#ux
“Engineers bring skills to the discovery process that other members of the product team may not have. This is why it’s so valuable to have them participate—even if they'd rather be writing code.” - @ttorres https://t.co/RK8F2OrKbp #prodmgmt#ux
"Healthy skepticism
Analysis as team sport
Measure to learn vs. control/manage
Kill features/experiments
Healthy approach to grappling w/uncertainty
High levels of confidence when safely experimenting
Measurement 'just is'"
by @johncutlefish https://t.co/ARowD2gJL4