Every company should be working on something like this 👇
Unlike humans, agents read all the fine print and will choose / recommend products that let them get more done
This is not always aligned with how humans see or use products, so it’s important to build for both
Over 1 million Granola notes are read by agents. Every day.
So we built a version of Granola for them too.
Agents now get expanded notes via MCP. With extra context – like explaining who Sam is or the latest news on Project Sprout.
So they're always up-to-date with your work.
Over 1 million Granola notes are read by agents. Every day.
So we built a version of Granola for them too.
Agents now get expanded notes via MCP. With extra context – like explaining who Sam is or the latest news on Project Sprout.
So they're always up-to-date with your work.
When we designed Granola, talking to users was our secret weapon. Sam and I would do 5+ interviews a week.
Now we do the same thing with agents 🤖. We interviewed tons of agents, asked them what they didn’t like about our MCP, and redesigned it to make them happy.
Introducing Granola MCP 2.0: the most efficient and intelligent way to work with your meeting context.
🔎 Smart Search: 85% fewer tokens. Transcripts stay the source of truth.
✍️ Write Tools: Add agendas and organise notes.
🧠 Agent Brief: Keep your AI up to date on your work.
Introducing Granola MCP 2.0: the most efficient and intelligent way to work with your meeting context.
🔎 Smart Search: 85% fewer tokens. Transcripts stay the source of truth.
✍️ Write Tools: Add agendas and organise notes.
🧠 Agent Brief: Keep your AI up to date on your work.
Dunno about you, but I hop between Claude, ChatGPT, Grokbot, Cursor, etc every day. No longer do i have to rely on their memories, or teach them what I know.
Plug in Granola and give them an always-up-to date picture of you to help them work better
The Big Labs are hoping that once you set up your agent, it’ll be too painful to switch. That’s why they let you connect apps via MCP, but don’t offer an MCP (even though they invented it 🤔).
We’re all better off when we’re not locked-in to one agent. Check out your agent brief, they are surprisingly good!
The Big Labs are hoping that once you set up your agent, it’ll be too painful to switch. That’s why they let you connect apps via MCP, but don’t offer an MCP (even though they invented it 🤔).
We’re all better off when we’re not locked-in to one agent. Check out your agent brief, they are surprisingly good!
Introducing Agent Briefs. Every night, Granola crunches everything it knows about you and summarizes it for your agents: priorities, habits, relationships - even how you communicate.
This way, all agents have your best, most up-to-date context.
See what yours says below 👇
Claude: grano.la/brief-claude
ChatGPT: grano.la/brief-chatgpt
The human brain is approaching its ChatGPT moment.
I spent 6 months writing Phase Lock: a manifesto on how brain computer interfaces can help align AI and preserve human agency.
If you only ever read one thing from me, read this.
The human brain is approaching its ChatGPT moment.
I spent 6 months writing Phase Lock: a manifesto on how brain computer interfaces can help align AI and preserve human agency.
If you only ever read one thing from me, read this.
The human brain is approaching its ChatGPT moment.
I spent 6 months writing Phase Lock: a manifesto on how brain computer interfaces can help align AI and preserve human agency.
If you only ever read one thing from me, read this.
I am very happy to share that our paper, “Ideological Alignment and Evidence-Based Policy Adoption,” is now published in the American Economic Review (#AER). This is a hugely ambitious project that has taken more than four years of work together with an amazing team of co-authors.
We conducted a nationwide #RCT involving over 5,000 real local politicians, as well as national think tanks, newspapers, and a leading global academic institution, to answer a simple but very important question: does the political “colour” of the messenger influence policy implementation?
One of the central challenges of the #evidence-based policy agenda is turning evidence into actual policy. From my experience in politics, this is far from straightforward: evidence has limited value if it does not translate into policies that improve people’s lives. Why is it so difficult to persuade policymakers to adopt policies with strong empirical backing? If the evidence is there, why are these policies so rarely implemented?
🔬 In our study, we address this by sending identical information about an evidence-based policy to local policymakers, introducing variation only in the #messenger of the policy: a non-ideological research institution versus think tanks and newspapers with clear ideological leanings. The policy itself is a recommendation - strongly supported by evidence- that improving a municipality’s Wikipedia page increases tourism.
We find that policy adoption increases by 65% when the information comes from a source perceived as ideologically aligned with the policymaker. In contrast, ideologically opposed institutions have no effect.
Publishing in what is arguably the world’s leading academic journal is a dream for any economist. For me - having stepped away from academia for politics - the joy is even greater.
🙏 I am deeply grateful to all my incredble co-authors for their generosity of offering me the opportunity to share this project with them, and for everything I have learned throughout this process. I also want to thank all the collaborators and RAs that have contributed, as well as across institutions, including eldiario.es, El Mundo, Alternativas, FAES, and FEMP, as well as the support from the Social Research Call of La Caixa and the Wikimedia Foundation.
aeaweb.org/articles?id=10.12…@IEuniversity@iespega@EnricoLetta@manuelmunizv@Borjamonde@JuanSantalo
Two weeks ago I vibe coded 7 AI products in 7 days.
Today I'm handing them all over to a OpenClaw agent called Clawrence to run autonomously.
Follow him below to see how he does!
I vibe coded an art project.
Founders put in the URL of their business. And it turns it into a weird and wonderful story about how their product was SO SUCCESSFUL it ended up destroying the world.
Because in the age of AI, maybe art is all humans will have left?
Link in the comments 👇