Pinned Tweet
What happens when AI agents have access to your money?
Who controls what they can buy, how much they can spend, and which services they can access?
@i_zerbib is the Founder & CEO of @sapiom, which has raised $50M in 2026 to build the payments infrastructure for AI agents.
Sapiom enables AI agents to independently purchase models, APIs, compute, and other services while maintaining security, authorization, and spending controls.
I recently sat down with Ilan on the @AIVenturesPod to discuss some of the bigger implications of this shift:
• Security and trust: How do you give AI agents the ability to transact autonomously without giving them unrestricted access to your money?
• The future of SaaS: What happens when AI agents, rather than humans, become the primary users and buyers of software, and how does that accelerate the shift from traditional seat-based pricing to consumption-based models?
• Building a new category: How to approach go-to-market and land your first customers through founder-led sales when you're bringing an entirely new category of technology to market.
• Fundraising and VC partnerships: How Ilan raised a $15M seed round and $35M Series A within the span of 6 months, and what he looks for when choosing the right venture capital partners.
👇 Check out the full episode through the link in the comments, and if you enjoy it, consider liking the video and subscribing 🔔 for more conversations with venture capital-backed AI founders.
🎙️ Watch the full conversation with Ilan Zerbib, Founder & CEO of Sapiom: youtu.be/v1ypZCbLtow
📩 Subscribe to the AI Ventures Newsletter for insights on enterprise AI, GTM strategy, and startup growth from conversations with venture capital-backed AI founders: linkedin.com/build-relation/…
1
23
Adam Knees retweeted
We’re launching the AI Ventures Newsletter!
Each edition breaks down insights from conversations with venture-backed AI founders on building, scaling, and growing their companies.
Our first edition covers how Stack AI built an enterprise AI platform and was acquired by Asana for $75M.
From finding product-market fit to navigating enterprise AI adoption, here are 4 lessons every AI founder should know
Subscribe on LinkedIn 👇linkedin.com/build-relation/…
Adam Knees retweeted
Replying to @retellai
@retellai reportedly hit $100M in annualized revenue with just $4.6M in total funding.
In an industry obsessed with billion-dollar funding rounds, this might be one of the more impressive startup stories.
Are we entering an era where small AI teams can build massive businesses?
I used to make fun of my dad for always talking to his phone.
Now I never type anymore.
Damn, life’s crazy
2
71
The Fireworks AI breach should be a wake-up call for every company building with AI.
Notion AI chats exposed. Customer datasets compromised. Even partial model weights accessed.
The bigger concern isn’t just model security. It’s how much sensitive enterprise data we’re trusting third-party AI providers with.
What do you think needs to change about how enterprises approach AI security?
techcrunch.com/2026/11/12/cy…
AI research labs: Stop worrying about GPU infrastructure and focus on building better models.
Snowflake is looking to partner with select well funded AI startups training and fine-tuning open-weight models.
We’re offering access to GPU compute, a model training environment, and direct collaboration with our Cortex Training team, including research engineers behind DeepSpeed and ZeRO.
Beyond infrastructure, this is an opportunity to help shape our product roadmap and accelerate your model development.
If you’re building in this space, shoot me a DM. Would love to connect.
Customize AI to your own data and context. Let Snowflake manage the GPUs.
Cortex Training, now in public preview, lets you post-train open-weight models with supervised fine-tuning, reinforcement learning and custom training loops.
Use frameworks such as @huggingface TRL, veRL and SkyRL, compare checkpoints and deploy through OpenAI-compatible endpoints.
Details: bit.ly/4t2I8Xt
Adam Knees retweeted
$75M Founder Explains the Biggest Distraction for Startups
#startup #entrepreneurship #business #founder #startups
One of the biggest mistakes startup founders make is getting distracted by what everyone else is doing. Bernard Aceituno, co-founder of Stack AI, which was acquired by Asana in a reported $75 million deal, explains why obsessing over competitors and market trends can actually hold your business back. Instead of constantly worrying about what other companies are building, founders need to focus on understanding their customers, identifying what works, recognizing what doesn't, and making their own products better. It's easy to convince yourself that you know what the market wants, but those assumptions can become one of the biggest obstacles to building something people genuinely need. Sometimes the smartest thing a founder can do is stop watching the competition and start listening to their customers.
The biggest shift as a founder isn’t working harder. It’s changing what you spend your time on.
As your business scales, you earn the ability to focus on higher-level execution: hiring, customers, and product strategy rather than juggling code, GTM, finance, and a million other hats.
That’s what scaling a company actually looks like.
Mercor just raised a $750M Series D at a $20B valuation
Here’s CEO @BrendanFoody’s daily routine:
› Wakes up around 7:45
› Works out most days ~7:50–8:30
› In the office by 9
› ~4 hours of meetings a day
› Spends a third of his time on hiring, a third with customers, a third on research & product
“I’m in the office until 10pm, & I rinse & repeat 6 days/week. And then Sundays I work from home.”
“I still inbox zero every day because I want to make sure I get through everything.”
I like Alexandr Wang, but I think he’s overlooking something here.
AI may make intelligence and execution abundant, but that doesn’t mean the barriers to building a successful business disappear.
Especially in B2B enterprise, where trust, relationships, distribution, and navigating complex buying decisions still matter.
I think he’s onto something at a smaller scale, but we’re a long way from AI replacing what it takes to build and sell into large enterprises.
.@alexandr_wang (Chief AI Officer of Meta) on why vision and ambition will soon become scarce:
"I think a decade looking back, it'll be obvious that intelligence became abundant and that agency became abundant. The current trends we're on are just going to keep continuing, and this will be very strange. For the history of humanity, groups of smart people getting together towards a shared goal was the bottleneck of progress.
The United States of America, in some sense, was an example. The United States of America is formed from a group of very smart people getting together and having a vision for the future that they wanted to enact. And that's the story of nearly every company in America and the story of every YC company.
And that's going to change. All of a sudden, the scarce resource isn't going to be intelligence or agency. I really think it's going to be vision and ambition.
Do you have a clear view of what you want the world to look like in the future? What is the one way in which you want to put your finger on the scale for how the future of the world will develop? And how will the world look in five to ten years in a way that it does not look like today? And you have the ambition and drive to go through all the crap to make that happen. And AI will make that easier. Agents and AI make that maybe 10 times or 100 times easier than it was a decade ago."
Video Credit: @ycombinator
Something about this whole Jensen Huang situation isn’t adding up 🤨
Months ago, Trump was easing restrictions on Nvidia chip sales to China. Jensen has consistently advocated for greater U.S.-China cooperation in AI.
Now Jensen says he had no idea he was involved with a suspected Chinese spy, Nvidia’s board launches an independent investigation, and Trump and Elon are defending him?
Makes you wonder how much of this is about national security versus protecting America’s economic interests in AI
Adam Knees retweeted
AI Is Changing the Job Market Forever
#ai #artificialintelligence #tech #chatgpt #jobs
AI is changing the job market forever, but not necessarily in the way people think. Stack AI co-founder Bernard Aceituno explains how artificial intelligence is reshaping the skills employers value and the work humans will be responsible for. As AI takes over more repetitive and analytical tasks, workers may need to become more adaptable, resourceful, and comfortable managing AI-powered workflows. But as these changes accelerate, people who don't fully understand AI may become increasingly hesitant to enter certain careers. The future of work isn't just about AI replacing jobs. It's about fundamentally changing what it means to be employable.
Your proprietary data shouldn’t be the price of admission for using frontier AI.
One of the biggest concerns I’m hearing from AI companies in highly regulated industries is exposing sensitive data to third-party model providers.
This is where Snowflake is uniquely positioned.
Through Cortex AI, companies can access models from 14 providers, including OpenAI, Anthropic, Google, Meta, xAI (Grok), DeepSeek, and other open-weight alternatives, all within Snowflake’s governed environment.
The result? Greater model flexibility, stronger data security, and the ability to optimize AI spend without sacrificing control over proprietary data.
🤖 Made with AI
For AI startups training their own models, securing GPU capacity is only half the battle.
Having the right training environment and GTM support is what separates building a great model from building a successful AI company.
@Snowflake is looking for a select group of well-funded AI startups to collaborate with our Cortex Training team through a design partnership.
This is an opportunity to work directly with our AI product engineers, including talent from Google and Meta, while gaining access to GPU capacity, a model training environment, and joint GTM opportunities through Snowflake's enterprise ecosystem.
If you're building a fast-growing AI startup in the Bay Area and this aligns with your technical roadmap, shoot me a message. Would love to connect.
Cloudflare confirming that agents drive more internet traffic is proof that this addressable market is going to to continue to grow exponentially so makes sense
Chamath Palihapitiya just made a pretty incredible comparison about Anthropic, and it puts into perspective how quickly the AI industry is growing.
He believes Anthropic could generate more cumulative revenue in its first seven years than three of the most successful businesses in history, Apple's iPhone, Google's search advertising business, and Amazon. COMBINED. The iPhone transformed the smartphone industry and became one of the most profitable consumer products ever created. Google built an advertising empire around search, essentially monetizing the world's information. And Amazon completely transformed e-commerce while building one of the largest cloud computing businesses in the world. These businesses took years, and in some cases decades, to reach their current scale. Anthropic, founded in 2021, is potentially approaching a comparable early revenue trajectory in a fraction of that time.
And the numbers coming out of Anthropic are pretty remarkable. Revenue grew from just $386 million in 2024 to $4.59 billion in 2025, and the company generated another $16.29 billion in the first six months of 2026 alone. That's an extraordinary acceleration for a company that barely existed five years ago. But what's even more interesting is how early we could still be in the adoption cycle. During the All-In discussion, they estimated that Anthropic has penetrated only around 1% of its potential enterprise customer base. If that estimate is anywhere close to accurate, the overwhelming majority of businesses haven't fully integrated AI into their operations.
And this is where the economics get fascinating. Unlike traditional software, where companies generally pay for a fixed subscription, a substantial portion of AI revenue comes from actual usage. Every time a company runs an AI agent, processes documents, writes code, analyzes data, or automates a workflow, it consumes tokens. As AI becomes more capable, companies won't necessarily just add more users. They'll begin assigning increasingly complex tasks to AI systems, allowing agents to operate continuously across entire business operations. That could dramatically increase the amount of computing consumed by each customer. And consider what happens as AI models become cheaper. A task that costs $10 today might eventually cost pennies, making it economically viable to automate millions of tasks that businesses wouldn't even consider automating right now. Lower prices could actually accelerate total consumption rather than reduce it, because the number of use cases grows so dramatically.
Way too many AI founders are niching themselves into oblivion.
A repeatable business needs two things: repeatable demand and a repeatable way to supply it.
“AI agents that optimize the menu pricing of vegan food trucks in Austin” might be a great demo. It’s probably not a venture scale company
Naval Ravikant on why founders should focus on lean startups:
“Steve Blank, who teaches at Stanford and started Epiphany among many other companies, he defines a startup very nicely. He says a startup is a search for a scalable and repeatable business model. And so what you're really doing is you're searching. And until you've found that business model that you can repeat and you can scale, you should stay very, very small and very, very cheap.”
If you ever start thinking you’re important, hike to the top of a hill.
You’re just an ant floating on a rock through space.
The difference @sama is talking about isn’t ambition. It’s ownership.
I talked about this with Stack AI founder @BernAceituno.
The same type of person who becomes a great investment banker could potentially become a great founder.
But some people would rather own the outcome than optimize for the path everyone else tells them is successful.
Sam Altman almost ended up as an investment banker before he became the founder of OpenAI
While he was at Stanford, every smart kid around him wanted to go into banking.
Sam Altman, feeling the peer pressure, recruited during his sophomore year and landed an internship at Goldman Sachs
He ultimately ended up turning down the internship and focusing on startups instead
Adam Knees retweeted
Replying to @AarushYadav47
@AarushYadav47 come on the podcast I wanna try these
We’ve raised $1.3M from YC & Afore Capital so you never have to type or speak again.
Typing is slow. Talking into your computer is awkward. Subvocal fixes both
Our glasses give you the speed of speech without you making a sound or touching a keyboard.
It tracks your internal speech muscles with non-contact sensing and decodes silent, almost invisible mouthing into text at more than 200 words/min.
Pre-orders are now live at:
subvocal.co/preorder?utm_sou…












