Common Magic

So fun to join @Kieranleehill #20Angel pod for frank chat on angel investing + my new fund @_CommonMagic If there's one thing I've put my 10k hours into, it's answering; "what IS community?" Listen + tell me what you think of my current definition 🔥 open.spotify.com/episode/6kw0K8…

my cousin pitched me a greeting card company 5 times over 5 years. i said no every single time. then in 2025 i finally said yes. to a paper company. while every founder i knew was racing to build AI agents, we walked into VC meetings pitching snail mail. we kept telling investors: in a few years saying your company "uses AI" is going to sound as ridiculous as saying you're "an internet company." so we skipped that part. escargot is not a greeting card company. it's a way to stay close to the 15-20 people you actually care about. we remind you when the moments are coming, make it stupidly easy to show up for them, and close the loop so you actually feel it. no algorithms. no hashtags. just slowing down and being thoughtful. today @BusinessInsider published a feature on what we're building. we raised $2.75M led by @NWischoff and @HannahGreyVC with @southpkcommons , @_CommonMagic , @NextWaveNYC , @bentossell , and @LizaGurtin . grateful to my co-founder and cousin Andrew for not giving up after 5 rejections. and to our team — this doesn't exist without you. cards are just the beginning 🐌

Piloted a breakfast for people figuring out what to build next with the incredible @sarahdrinkwater at the @meetgranola office. Early feedback's been good, so likely doing this again. If you're exploring what to build next and want in, lmk! 👋

We spent two years getting LLMs to speak valid JSON. That was the easy part. Now we're solving what actually matters: making sure the model can't even express an action it's not allowed to take. Not "will not." Cannot. 👇 blog.dottxt.ai/control-layer-…

Trust Everything, Everywhere Hackathon, 7-8th March | @sotalikesfuture x @ARIA_research How do we make future cyber-physical systems safe, secure and robust under adversarial constraints? In other words, how do we scale trust infrastructure to the physical world? In March, SoTA is organising a hackathon to start answering these questions. We’ll bring together a group of earnest, high-agency and industrious individuals to explore making AI deployment trusted in an increasingly programmable cyber-physical reality. Multi-disciplinary approaches are strongly encouraged. We want to see what happens when an applied cryptographer, a network defender and a computational biologist collaborate on securing a drone swarm, or an air traffic control system, or a self-driving biolab. Our ambition is to redirect more talent towards building the security primitives – such as advanced cryptography and hardware verification – needed for trusted cyber-physical interactions enabling physical AI adoption at scale. Cash and compute credits will be awarded for the Grand Prize, Technical Excellence Prize, and Novel Approach Prize. Register at the link and read our letter to friends of SoTA below. luma.com/eron45y6

The Moltbook stuff is still mostly a nothingburger if you've been following things like the infinite backrooms, the extended Janus universe, Stanford's Smallville, Large Population Models, DeepMind's Concordia, SAGE's AI Village, and many more. Of course the models get better over time and so the interactions get richer, the tools called are more sophisticated and so on. I'll concede that at least it's making multi-agent dynamics a bit easier to understand for people who are blessed with not spending their days interacting with models and monitoring ArXiv. The risk side is easy to grok - it always is! Humans are very good at freaking out. And whilst I like poking fun at the prophets of doom and the anxiety/neuroticism fueled parts of the AI ecosystem, it's plainly true that safety is important. So it's a good time to remind people of the Distributional AGI Safety paper (arxiv.org/abs/2512.16856) and the Multi-Agent Risks from Advanced AI paper (arxiv.org/abs/2502.14143). There's a lot to research here still. As usual, this will benefit from people with deep knowledge in all sorts of domains like economics, game theory, psychology, cybersecurity, mechanism design, and many more. Maybe this is the year we will get better protocols to incentivize coordination and collaboration without the downsides, mechanism design and reputation systems to discourage malicious actors, and walled gardens and proof of humanity to better filter slop. And risks aside - I think there's so much to be researched to help enable positive sum flywheels: using agents to solve coordination problems, OSINT agent platforms to hold power accountable, decentralised anonymized dataset creation for social good, aggregating dispersed knowledge without the usual pathologies (Community Notes for everything!), simulations of social and political dynamics, multi-agent systems that stress-test policy proposals, contracts, or governance mechanisms by simulating diverse strategic actors trying to game them etc. It's time to build!