Weekly Dose of Optimism #138
Isomorphic Labs, OpenAI Round, Streaming Thoughts, Dementia Vaccine?, Lipoprotein(a), Hyperlegible 005
Hi friends 👋,
Happy Friday and welcome back to our 138th Weekly Dose of Optimism. We hope that your week was tarrific and not too tarrifying. Instead of staring at your portfolio all weekend, may we recommend you indulge in some slightly more optimistic news to help ease the pain?
Let’s get to it.
Today’s Weekly Dose is brought to you by… Vanta
Vanta automates up to 90% of the work needed for SOC 2, ISO 27001, and more to get you compliant fast. Vanta opens the door to growth.
Vanta is one of the world’s fastest-growing startups, because whether you’re closing your first deal or gearing up for growth, Vanta makes compliance easy.
It’s time to grow your ARR. Vanta can help Get $1,000 off Vanta here:
(1) Alphabet’s AI Drug Unit Raises $600 Million From OpenAI Backer
Dana Wollman for Bloomberg
Isomorphic Labs, the Alphabet Inc.-owned company that uses artificial intelligence to discover drugs, has raised $600 million, the first time it’s taken in external funding.
AI is making money moves into pharma. Alphabet’s drug discovery startup, Isomorphic Labs, just raised $600M in its first-ever external funding round—led by Thrive Capital, a major backer of OpenAI. The smart money sees AI not just as a chatbot engine, but as a future drugmaker.
Founded by DeepMind CEO Demis Hassabis, Isomorphic aims to accelerate how we discover and develop new medicines. Backed by groundbreaking work in protein folding and now partnered with Novartis and Eli Lilly, the company is laying deep foundations. It hasn’t produced a blockbuster yet, but it's quietly building the infrastructure to do so.
AI is moving upstream into biology, and the companies with the deepest models and pockets may soon shape the future of medicine.
(2) New funding to build towards AGI
From OpenAI
Today we’re announcing new funding—$40 billion at a $300 billion post-money valuation, which enables us to push the frontiers of AI research even further, scale our compute infrastructure, and deliver increasingly powerful tools for the 500 million people who use ChatGPT every week.
I have a hot take. Or at least, in the context of our current economic policies, it's a hot take. Competition is good for consumers. Competition drives down prices and drives up quality and spurs innovation. Which is why I love what’s happening in AI right now. It’s competition on steroids, and for the most part, it’s all happening right here in the US.
When OpenAI raises $40B, that’s good for consumers. It’s also good for OpenAI, to be clear. But my perspective is that it’s ultimately even better for the consumer. It increases competition. Its larger competitors, like Google or Amazon or Microsoft, may look at OpenAI now and say, “hmm they seem like an even bigger threat, let’s make all of our really good products completely free for consumers and try to slow their growth.” Its smaller competitors, like Anthropic or more verticalized AI players, are probably like “ohh shit, no way we can’t compete on capital with them, we just really to need increase our product quality and innovation.”
It’s unclear where the value will ultimately accrue amongst these companies. But what is certainly clear is that consumers are going to get cheaper, faster, better AI products thanks to this period of AI company knife-fighting.
(3) A streaming brain-to-voice neuroprosthesis to restore naturalistic communication
Littlejohn et al, h/t Vittorio
Our findings introduce a speech-neuroprosthetic paradigm to restore naturalistic spoken communication to people with paralysis.
We can now stream thoughts to voice.
A breakthrough in neural speech prosthesis enabled people with paralysis to speak silently and fluently just by thinking. A new deep learning model decodes brain activity into real-time speech at up to 90 words per minute (WPM), which is triple the previous record. It then synthesizes the user's own voice and works continuously, without prompts or delays. Even more impressively, it decoded words never seen in training.
Unlike past systems, it decodes pure motor intent instead of it relying on sound or audio feedback, which is a big leap toward thought-to-speech communication. No need to vocalize, no keyboards, just pure thoughts → language. How this technology will evolve is anyone’s guess: wearables, use cases outside of the disabled community, no one ever actually speaking again…who knows? But this is the type of SciFi-y, Telepathy Tapes-esque breakthrough we absolutely love to cover in the Weekly Dose.
(4) A natural experiment on the effect of herpes zoster vaccination on dementia
Eyting et al in Nature
Through the use of a unique natural experiment, this study provides evidence of a dementia-preventing or dementia-delaying effect from zoster vaccination that is less vulnerable to confounding and bias than the existing associational evidence.
The shingles vaccine may delay dementia. That’s according to new, clever research coming out of the UK which found that people just barely eligible for the herpes zoster (shingles) vaccine were 20% less likely to be diagnosed with dementia over the next seven years.
The researchers took advantage of a strict age cutoff — those born just one week apart had vastly different vaccine rates but were otherwise identical, allowing for unusually strong causal evidence. Back when the Welsh government rolled out the zoster vaccine in 2013, it made people who turned 80 before September 2, 2013 permanently ineligible, while those who turned 80 on or after that date were eligible for one year. This arbitrary line created a natural “randomization”: people born just days apart had a 47% difference in vaccination rates, enabling researchers to isolate the vaccine’s effect without typical biases.
The study’s findings support the theory that either herpesviruses themselves contribute to dementia or that the immune activation from live vaccines like Zostavax provides off-target neuroprotection. Women and people with autoimmune conditions benefited more, hinting at immune system mechanisms. Importantly, this effect wasn’t seen with other diseases, vaccines, or behaviors, ruling out confounding.
(5) Lepodisiran — A Long-Duration Small Interfering RNA Targeting Lipoprotein(a)
Nissen et all in the New England Journal of Medicine
Lepodisiran reduced mean serum concentrations of lipoprotein(a) from 60 to 180 days after administration.
You may have never heard of lipoprotein(a), but it can kill you. It’s linked to a 2x higher risk of heart attacks and strokes for 1 in 5 people. The genetically determined particle is made of cholesterol, fats, and a sticky protein called apolipoprotein(a). High levels can clog arteries and dramatically increase the risk of heart attacks and strokes, especially in people with no other clear risk factors and, frustratingly, it’s unaffected by diet, exercise, or most existing drugs. Doctors don’t even screen for it because there’s not much they can do about it.
Or at least we thought so!
Lepodisiran, a new injectable from Eli Lilly, reduced lipoprotein(a) by up to 94% with a single dose, and the effect lasted six months. The early trial showed dramatic biomarker reduction and no serious side effects. Yet, it’s ability to prevent heart events is still unproven. Large outcome trials are ongoing, with Eli Lilly's wrapping in 2029 and Novartis leading a faster path with results in 2026. If outcomes match biomarkers, this could be the statin moment for genetically-driven heart disease.
We like telling cancer to go fuck itself in the Dose, but it feels good to say it to heart disease, an even bigger American killer, too. So get fucked, heart disease.
Bonus: Hyperlegible 005: Alex Danco, Scarcity and Abundance in 2025
Packy here. I’ve been having a ton of fun interviewing the people behind my favorite essays for Hyperlegible over the past two weeks. Tina He, Utsav Mamoria, and Julian Lehr have all exceeded my already-high expectations.
Now comes Alex Danco. Alex has been one of my favorite internet writers since before I became an internet writer. I wrote about his essay, Positional Scarcity, way back in April 2020s Wackos and ZoomGlüts. He’s appointment reading for me.
So I had high hopes for our conversation on his new piece, Scarcity and Abundance in 2025, and he… completely demolished them. He’s just so good. One of those guests that you could throw three random words out there and he’d figure out what you were going for and give a much better answer than you could have ever anticipated.
The essay, and therefore the conversation, is about the dynamic between scarcity and abundance, where value accrues when new abundance (and therefore new scarcity) is created, how companies like Apple, Cursor, NVIDIA, and YC might fare, and how Agents + Crypto might actually become a thing.
Very few people have shaped how I think about tech companies as much as Alex. Listen to our conversation to understand why, and then go devour his back-catalog.
Have a great weekend y’all.
Thanks to Vanta for sponsoring. We’ll be back in your inbox next week.
Thanks for reading,
Packy + Dan
Thoughts to speech?? That is scary. Here come the Dream Police like in the Cheap Trick song.
Interesting news about the shingles vax.