attorney, innovator, investor legal tech background
Joined July 2023
- Tweets1.4K
- Following365
- Followers107
- Likes2.2K
A. retweeted
Systems > goals
My best advice is to stop using motivation as your only fuel. I know it feels great when you’re fired up, but it’s a short-term fuel source. That’s why the vast majority of people who start anything - diet, fitness, new projects - don’t finish. They run out of gas.
The only lasting fuel is routine. And you only get a routine by dragging yourself on the days when you have no motivation. Over and over.
I know that’s not the answer anyone wants. I wish I had a magic pill for you. But the only thing that works long term is showing up for yourself even when you don’t want to. Brute force.
I’m slightly crazy and don’t have any investors or consultants to listen to, so I’m giving people 50 dollars of their 100 dollar annual subscription back when they show up for themselves and complete a full program in my app. If that motivates you to start, it’s designed to build your routine to keep you going like it has for thousands of other people, so join us: arnoldspumpclub.com/pages/be…
The type of robot I love to see:
A robot is cleaning rivers completely autonomously. 🌊
Meet Hebao H3, an autonomous surface-cleaning boat developed by Dongfang Water.
Using onboard AI vision and GPS, it can detect and collect floating waste such as plastic bags, bottles and leaves while avoiding obstacles.
When its collection bin is full, the H3 can automatically return to its base.
Would you trust robots to keep our rivers clean? 👀
⚠️ This content is shared for informational purposes only. CTO Robotics Media is a media platform and does not own or develop the technology shown. Credit belongs to the original creators.
#HebaoH3 #DongfangWater #RiverCleaning #Robotics #AutonomousRobots #AI #EnvironmentalRobotics #CleanTech #Sustainability #Automation #CTORoboticsMedia
Baseball should be played outdoors.
#forthefaithful
A. retweeted
Confirmed: @Muse can sue people in small claims court
$WMT playing it smart…
All of my Muse purchases have been through Walmart. Every time it was the lowest cost. I never would have considered browsing Walmart in the past.
A. retweeted
The most meaningful relationships are achieved when you and others can speak openly to each other about everything that's important, learn together, and understand the need to hold each other accountable to be as excellent as you can be.
When you have such relationships with those you work with, you pull each other through challenging times; at the same time, sharing challenging work draws you closer and strengthens your relationships.
This self-reinforcing cycle creates the success that allows you to pursue more and more ambitious goals. #principleoftheday
Even still, it is time to build!
pmarca.substack.com/p/why-ai…
What Is To Be Done?
I propose a simple plan:
Big AI companies should be allowed to build AI as fast and aggressively as they can – but not allowed to achieve regulatory capture, not allowed to establish a government-protect cartel that is insulated from market competition due to incorrect claims of AI risk. This will maximize the technological and societal payoff from the amazing capabilities of these companies, which are jewels of modern capitalism.
Startup AI companies should be allowed to build AI as fast and aggressively as they can. They should neither confront government-granted protection of big companies, nor should they receive government assistance. They should simply be allowed to compete. If and as startups don’t succeed, their presence in the market will also continuously motivate big companies to be their best – our economies and societies win either way.
Open source AI should be allowed to freely proliferate and compete with both big AI companies and startups. There should be no regulatory barriers to open source whatsoever. Even when open source does not beat companies, its widespread availability is a boon to students all over the world who want to learn how to build and use AI to become part of the technological future, and will ensure that AI is available to everyone who can benefit from it no matter who they are or how much money they have.
To offset the risk of bad people doing bad things with AI, governments working in partnership with the private sector should vigorously engage in each area of potential risk to use AI to maximize society’s defensive capabilities. This shouldn’t be limited to AI-enabled risks but also more general problems such as malnutrition, disease, and climate. AI can be an incredibly powerful tool for solving problems, and we should embrace it as such.
To prevent the risk of China achieving global AI dominance, we should use the full power of our private sector, our scientific establishment, and our governments in concert to drive American and Western AI to absolute global dominance, including ultimately inside China itself. We win, they lose.
And that is how we use AI to save the world.
It’s time to build.
A. retweeted
Thomson Reuters announced it is moving off Claude to Alibaba's Qwen to cut costs.
From today's Anthropic report: Alibaba extracted 151M+ Claude exchanges to help train Qwen.
TR now deploys Westlaw on a vast trove of stolen US IP.
@AnthropicAI's report: anthropic.com/threat-intelli…
A. retweeted
No matter what side you are on, the pace is set and nothing will stop it. It doesn't matter if it's Anthropic, OpenAI, Moonshot or others.
The question we should be pondering is how do we steer the outcome and what do we do when it inevitable comes, not how to slow things down.
A. retweeted
Courts are sorting AI into two bins: general models as transformative, purpose-built tools as substitutes. I think this line skips the actual fair use work, and I use TR v. ROSS to show why. And yes, I have a stake. I cofounded ROSS, the appeal is pending.
Even the lawyers with AI access are mostly squibs. Even the lawyers who use AI are mostly Gilderoy Lockhart.
A. retweeted
The Navier Stokes proof drama is good foreshadowing for what happens / becomes important in the entire economy going forward: spotting opportunities, pointing models at them, and validating solutions faster than others. In other words, the economy / knowledge discovery increasingly looks like quant trading / pod shops / HFs, where having a differentiated view of why an opportunity is valuable and/or realizing that an opportunity is valuable (and that it can be plausibly captured/exploited) faster than others is all that matters.
This is from a piece I wrote in October 2025 and I think it will continue to hold a lot of explanatory power.
We congratulate Levent Alpöge and Tristan Buckmaster on their remarkable mathematical work.
We (the researchers and the agents) did not see any of their work through any means until they released it publicly — in particular, no specific user data was accessed in order to solve this problem.
While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models.
However, our proofs differ significantly and even the precise results proved are different in the Euler case (forced vs. unforced).
According to @WeAreLegora, "GPT‑6 Astra improved performance by nearly 40% over the previous model" on its financial-statement workflow.
openai.com/index/legora-fina…
A. retweeted
“The highest form of peace is having no desire to be seen, validated or understood by anyone.”
— Epictetus
A. retweeted
Response from Thomson Reuters research head. What do you all think?
Thomson Reuters just announced their own post trained model Thomson-1 which is post trained on Qwen.
A caveat is that the benchmark in the press release compared the Thomson model with an outdated models including Gemini Pro 3.1 which was released more than half a year ago. And ChatGPT 5.5 had its reasoning turned off for some reason so it was the worst performing.
But the point is not about having the best model in the market that is competitive with the frontier but a good enough model for some legal workflows.
The reality is that it’s more of a cost control measure which makes sense. Simply using a Chinese OSS model is a hard sell to US corporate clients.
Notably the model will be deployed for Tabular Analysis which is the bulk summarisation of large document sets. I have heard that this feature is a big cost drain for Harvey and Legora as well. Doc review teams will put in hundreds of documents and run them in parallel. I’m sure Thomson is going to be more than enough for summarisation and will help TR save some money on tokens.
This is also a very interesting data point on large American companies shifting to using Chinese OSS models. Notably in 2024 TR worked with OpenAI to post train o-1 but that was never put into production. We are seeing a gradual decentralisation of value capture in the AI space
