Neal Mohan

The Headline family is experiencing incredible momentum, and today our portfolio company DeepIP is announcing a $25M Series B, bringing total funding to $40M.  This is the kind of company that doesn’t just add AI to a workflow, but it sets the standard for an entire category. The momentum speaks for itself: 1️⃣ 10x ARR growth in < 18 months 2️⃣ Trusted by 400+ law firms and leading corporate IP teams globally 3️⃣ 2M+ chats on the DeepIP platform We’ve had the chance to back FX (Francois-Xavier) L. and Edouard D., leading their Seed round with Headline. Watching them take on a problem as complex and high-stakes as patent work has been genuinely impressive. And working with them has been an absolute pleasure: these 2 are coupling very high ambition and humility. Tech built in Europe, but opening the US office very early and FX starting a life in NYC with his family with 0 hesitation in order to build a global category leader in the IP industry. What convinced me early is still what stands out today: patent teams don’t need another standalone AI tool. They need something that fits the way the work actually happens: documents, long-running matters, multiple stakeholders, and zero room for sloppy context handoffs. Big congrats to the whole DeepIP team + Korelya Capital, Balderton & Serena Proud that we’re doubling down at Headline and continuing the support for what’s next!

124

10 Comments

Love this - AI is the Oppenheimer moment for marketing. It fundamentally collapses the time and cost of production. A $2.5 million campaign that used to take four months can now be delivered for $500,000 in four weeks. Legacy firms that charge based on hours with the 'time-driven' model are existentially threatened by this.

23

Checked out Forerunner/Kirsten Green's "The Architects of Experience" piece (linked in comments). Sharing some thoughts below. Every 10-year platform shift resets consumer expectations (Web → Mobile → AI). Kirsten says the AI era belongs to products that feel effortless: proactive, predictive, always learning. It’s not about adding an LLM garnish; it’s about architecting the entire business around a compounding data loop. So, why does it matter? Moat ≠ code → The defensible layer is a proprietary data flywheel that learns faster than users can churn. Talent remix → Winning teams look less like pure engineers, more like “orchestrators” who weave off the shelf AI, data strategy, and distribution. Trust is the gate → Human in loop design and transparent governance are table-stakes for adoption, and eventual regulation. Questions to consider for consumer AI founders... 1. Moat depth: What makes your data flywheel unique, and how quickly does it compound vs. the big LLM platforms? 2. Distribution edge: When AI tooling is commoditized, what non-technical channel (community, brand, regulated access) gets you in front of users first? 3. Human-machine balance: Where does your product deliberately stop automating to preserve agency and delight? 4. Regulatory resilience: How will privacy portability rules affect CAC, margins, or data lock in? My counterview: Total automation can flatten delight. Some categories thrive on intentional friction (e.g. luxury, craft, serendipity). Winners will know when to keep a human in the loop.

33

4 Comments

We hosted Bret Taylor (Chair of OpenAI and Co‑Founder of Sierra) in San Francisco for a live taping of The Verge’s Decoder with Alex Heath. I had the privilege of co‑moderating a closed‑door roundtable on agentic AI and the future of enterprise software with industry leaders and investors. The discussion was candid on what’s working, what’s stalling AI at scale, and what leaders do next. Key themes: – AI tech is ready; organizations aren’t. Trust, governance, and culture are the real unlocks. – Agentic AI is exploding in internal use. In revenue‑critical customer workflows, leaders are far more cautious. – SaaS pricing models are under real pressure as AI reshapes value; buyers want more control, not surprises. – Auditability, transparency, and human‑in‑the‑loop design are non‑negotiable for agentic trust. – Executives and teams must become AI orchestrators. Upskilling is now a leadership responsibility. The Decoder conversation with Bret and Alex is live here: https://lnkd.in/gNjWuutD Grateful to Bret Taylor, Alex Heath, OpenAI, Sierra, and our AlixPartners team for making it happen.

121

4 Comments

This Scale AI story might matter more than you think. Meta just invested $14.3B to take a 49% stake in Scale AI — and brought Scale’s CEO into its internal “superintelligence” unit. On the surface, it looks like a major bet on data. But look a little deeper, and it’s a shift in how control is consolidating across the AI value chain. Why it matters: – Scale was one of the few trusted, neutral data providers for labeling and evaluation – Now that neutrality is in question, contracts are reportedly getting pulled by OpenAI and Google – Startups and enterprises alike are suddenly reconsidering where their training signal comes from What comes next: – A surge of boutique data providers — decentralized, domain-specific, and more transparent – Startups reframing data control as part of their moat: “We don’t just use models — we own the context” – Infra teams re-evaluating dependencies — either proactively, or reactively when vendor dynamics shift As training data usage becomes more regulated and controlled — moving away from the wild west of early foundation model training — the pressure to secure differentiated data pipelines only increases. When we talk about defensibility in AI, it’s tempting to focus on model IP or feature differentiation. But data leverage — how it’s sourced, labeled, evaluated, and governed — is becoming a more important part of the strategic equation. It’s not quite infrastructure. But it’s no longer a commodity either. It’s a layer of increasing consequence — and one worth watching closely. Are you rethinking your data leverage yet? https://lnkd.in/gx9GPefp

11

1 Comment

How do you turn AI uncertainty into ROI? Last quarter, a mid-sized NYC tech firm faced stalled innovation and decision fatigue. Their leadership team was overwhelmed by conflicting AI automation options. We stepped in with a cognitive audit, mapped their bottlenecks, and designed a private AI agent system tailored to their workflows. Within weeks, project delivery times dropped by 27%, and decision cycles became days, not weeks. Curious how a cognitive science-led approach can unlock your team’s next breakthrough? Let’s talk about your challenges. (If you’re navigating complex tech decisions, I’d welcome a direct message or connection.)

5

💭 𝗔𝗺𝗲𝗿𝗶𝗰𝗮 𝗶𝘀 𝗻𝗼𝘄 𝗼𝗻𝗲 𝗯𝗶𝗴 𝗯𝗲𝘁 𝗼𝗻 𝗔𝗜 Ruchir Sharma's Financial Times piece is insightful and on point—the U.S. economy has turned into a single, massive bet on AI. AI is now the default fix for everything—productivity, growth, and global competitiveness. Markets are pricing in perfection, with a handful of AI-proxy stocks driving >80% of the S&P 500 gains. I am long on AI’s transformative potential; however, when one technology becomes the primary growth engine for an entire economy, the concentration risk is real. 💠 The upside: faster adoption, cross-sector innovation, and genuine productivity gains. 💠 The risk: over-dependence, inflated expectations, and fragile optionality if AI ROI lags behind the hype. Leaders and investors need to balance conviction with caution—building real capabilities and growth & productivity engines, not just narratives. #AIacceleration #DigitalTransformation #Innovation #FutureOfWork

16

3 Comments

Love it or hate it, being in the SF Bay Area is still one of the biggest hacks for First Builders to discover huge opportunities before anyone else. But what can you do if you’re not here? Cecilia for example moved here in 2001 and instantly joined Yahoo after feeling the palpable energy around early internet. And it’s been nonstop ever since as she later joined Amazon, Cruise, Replit, and her own breakout company GC AI. She credits a lot to the emerging tech density and optimism in the Bay Area. But she also mentions additional ways people can identify these opportunities early on: - Follow intuition around big waves - Follow your envy - Never lose curiosity - Watch the teen hackers - Talk to peers early and often (builds long term networks/insights)

21

3 Comments

At Amazon, we worked on Prime Video and Amazon Music for 8 years before they were successful. We kept going because the total addressable market (TAM) is enormous— nearly every household on Earth listens to music and watches video. When deciding between several good new product or business ideas, pick the one with the highest TAM. Achieving success with a new product or business is hard. In most cases, it takes years of iteration, investment, and staying in the market until the timing is right. In contrast, coming up with new product or business ideas is relatively easy. At Amazon, we reviewed tons of new product and business ideas every year. Lots of these ideas were very good, but did not make the cut simply because the relative TAM was smaller than that of other ideas. In choosing between ideas, we always picked the ones with the biggest TAM. It takes the about the same amount of effort to start and manage a small business as it does a large one. So, using TAM as a factor in deciding what to focus on is one of the most valuable things that any business can learn from Amazon. You want all of the work to have a higher payoff in the end.

148

8 Comments

With the launch of Claude Co-Work (and tools like it), I feel like every week in 2025 carries the weight of a decade. Some thoughts: 1. The technology is genuinely different this time. Claude Co-Work isn't an incremental improvement — it's a 100x leap over what existed 12 months ago. If you haven't used it yet, stop what you're doing. 2. Markets. Software and advisory businesses face an existential reset .Not decline. Reset. Consulting, legal, accounting, financial advisory — these industries aren't disappearing, but they're heading toward 80% fewer people doing the same work. KPMG just told auditors to cut costs by 50%. This is the beginning, not the exception. 3. Real-world industries aren't safe either — they're just next. Industrials, energy, consumer goods, travel all have a natural (physical) moat. But consumer brands being built today are doing it with 80-90% less headcount than their predecessors. Any company older than 6 months that isn't adapting is already behind. Jack Dorsey's Block cutting 50% of its workforce despite strong growth is a preview, not an outlier. 4. The new bottleneck is energy, not labor. Once labor becomes abundant and cheap, the constraint shifts. Whoever can produce energy at scale and reasonable cost will have the most important infrastructure asset of the next 20 years. 5. The job market is being structurally changed for good and this is not a minor adjustment. Bankers, lawyers, accountants, software engineers — some roles will survive the way painters survived the printing press. Most won't. The talent that moves fastest onto new tools will capture a disproportionate share of the upside. We're entering a period of massive productivity unlocks and equally massive collateral damage. The future is bright. The transition is not going to be comfortable. Go test out some of those new tools.

114

19 Comments

Every publisher wants to prove their value — but how many truly understand how a CMO evaluates media partners? In this 20-minute conversation, I sat down with Lauren Weinberg (former CMO at Peloton and Square) to talk about what really drives media investment decisions. We cover: • What CMOs actually care about (and what’s just table stakes) • The role of MMM, MTA, and incremental ROI • A winning case study: Peloton × Twitch • Why impressions ≠ outcomes If you're in publishing or ad tech, this is one to watch. 🎥 https://lnkd.in/gMrWYuui

24

3 Comments

This week was so productive. We had a very productive work session with the AMD team, launching collaboration with Omeed Momeni & the University of California, Davis, followed by dinner and a great conversation with Mike Torre, a longtime Google veteran, former technical operations executive & advisor, and someone with deep experience scaling large, complex infrastructure environments. One of the main AI data center pain points was obvious. Modern data centers are increasingly constrained by unpredictable congestion inside the rack and across blades, where bursty, synchronized traffic from AI and distributed workloads creates microbursts and incast that overload fixed, wired switch paths in microseconds—driving queue buildup, tail-latency spikes, and, in lossless fabrics, head-of-line blocking and congestion spreading. Traditional networks are architecturally static: traffic is forced through pre-defined switch topologies that can only react after congestion forms, leaving operators unable to dynamically steer load away from hot spots or prevent cascading overload. RIVVOR Inc. Fabric fundamentally changes this by introducing a wireless, blade-to-blade, dynamically reconfigurable fabric layer that enables direct, on-demand paths between blades, absorbs burst traffic, and rebalances traffic in real time. The result is predictable, low-latency performance and high utilization precisely when conventional fabrics become unstable. #ai #aiinfruststructure #aidatacenters #datacenters #mmwave #amd #opencompute #ocp #wireless #infrustructure #hyperscalers

15

1 Comment

Even the most skeptical voices are starting to acknowledge the value of LLM generated code. Not as a replacement for deep expertise, but as a practical tool for acceleration, exploration, and iteration. When someone as famously disciplined about code quality as Linus Torvalds publicly acknowledges that AI generated code can be useful, it signals a shift worth paying attention to. The takeaway is not that AI writes better software than humans, but that good engineers are learning where and how these tools create leverage. https://lnkd.in/eFbNFqYN

6

Investors (esp deep tech investors) generally think about things/bet on "layers": the application, the infrastructure, the model. I think that's becoming the wrong way to look at it? The real opportunity is in the feedback loop *between* the layers. A new social app creates demand for new hardware, which in turn demands a new kind of intelligence - I'm calling it "The Agency Loop". My latest newsletter explores this thesis through three signals: the "weaponized transgression" of Cluey, the AI-driven materials science in Nature, and the architectural critiques of Kenneth Stanley. Bet the loop, not the layer :) You can read the full analysis here: https://lnkd.in/ggdVGEda #vibecap #vibecapital #vc #deeptech #AI #cluely

4

1 Comment

This is a really important take from Sundial & Chandra Narayanan on the METRICS TRAP: "Any individual metric, taken in isolation, supports multiple contradictory interpretations. The number itself doesn’t tell you whether the company is thriving or struggling. Only context does." https://lnkd.in/gxDWMYDV

25

1 Comment

What does it really mean to be building a Consumer AI-native product today? BBG Ventures continued its Breakfast Series with a convo on Consumer AI co-hosted with Factorial, Lightspeed, New Enterprise Associates (NEA), Stellation Capital (thanks SVB!) We were joined by two forward thinking AI-native builders - Dennis Crowley at Hopscotch Labs and Anna Bofa at Crate - and got into the hard stuff: what's valuable or just a feature; what does real personalization look like; what makes a product a no-brainer for user adoption; what are the new moats, and what are VCs looking for when they invest early? Our top takeaways below - weigh in with your thoughts! ✨ What’s newly possible vs. what’s newly trivial. AI makes a lot easier—but only some things more valuable. The best builders are distinguishing between products that unlocks new user behavior, that are meaningfully differentiated (possible) and features that simply make things easier - or could be built in a weekend (trivial). VCs are looking to invest in the former. H/t Matthew Hartman for this frame. 🧠 Is personalization the VIBE N of 1 personalization is the holy grail - but it's about more than prediction, it's about empathy. It's about building memory over time—capturing user tone, goals, and intent—and creating a vibe match that makes the product feel intuitive and emotionally aligned. That’s what builds trust, repeat use, and brand equity. Crate is doing this by collecting implicit signal, and Hopscotch's Beebot through its ambient "pop in your airpods" approach. Which brings us to ... ⚖️ The product "effort spectrum" Every product lives somewhere on the no → low → hard effort spectrum. AI-native experiences need to be closer to the "no-effort" end of the spectrum - a value exchange and clear payoff for little work. Put another way, friction is the enemy of frequency. 🧱 Moats are the same but different Fundamental moats remain the same, but network effects, proprietary data etc, but they get more exciting because of feedback loops and real time user data - for each interaction and related improvement. Hot take related to personalization from me: maybe the vibe match becomes the moat? 📊 What VCs are looking for at Seed vs. Series A At Seed, the VCs in the room Matthew Hartman Peter Boyce II and BBG Ventures are backing founders with a deep POV on behavior, training loops, and interface design. We prize organic growth and WOM. By Series A, investors like Michael Mignano Danielle Lay are looking for continued organic growth, strong retention and stable user cohorts, real engagement depth, customer obsession and proprietary data We're investing in Consumer AI! If you're building something newly possible and you believe in wow moments; if you’re working on n of 1 personalization and the vibe match with less work users come talk to us. Full post 👉 https://lnkd.in/geFief5D Claire Biernacki Susan Lyne

39

5 Comments

LLMs are about to disrupt search the way that search disrupted classifieds. Our brand new SVP Product, Teddy Jawde, is a guest writer on this week's Inside Performance Advertising newsletter. Click the link below to hear his thoughts on the 4 major shifts happening around the new monetization models native to the LLM experience.

52

3 Comments

Thrilled to announce that Together Fund is investing in Sentra, alongside a16z speedrun! You track results in Jira. Decisions in Notion. Conversations in Slack. But the reasoning, the debates, trade-offs, and context behind why you chose A over B, disappears into what we call "Dark Matter." A decision made in March looks insane by July because no one remembers the constraints that made it smart. I lived this firsthand at Twitter scaling from 800 to 8,000 employees, and at Google while launching AI Overviews to billions at planet scale. The problem isn't process. Process is compensation for something deeper: organizational amnesia. An organization’s "Systems of Record" doesn’t solve this, they encode it. They store what happened, never why. That's why we are investing in Sentra. Sentra is the always-on collective memory that eliminates organizational amnesia by maintaining accurate context for all members and agents, functioning as an operational nervous system. It connects to every channel where work happens, meetings, Slack, email, code commits, docs, calendars, and treats them not as artifacts to search, but as living signals to synthesize. The fleeting and the permanent, unified into a memory that understands. The founding team is built for this: - Jae Gwan Park (CEO): Product-first founder, memory systems research at UofT and MIT - Ashwin Gopinath (CSO): Former MIT professor, created "Reflexion" (NeurIPS 2023), agents that learn from mistakes, 2x founder - Andrey Starenky (CTO): Early Vapi engineer, ex-IBM, built to process enterprise-scale data firehose Together is an operator-led fund. We invest in problems we've lived. This is one of them. Many congrats Jae, Ashwin and Andrey, we are so excited to partner with you! Read the full thesis: https://lnkd.in/gixj9cE4 Book a demo: https://www.sentra.app/ #OrganizationalMemory #AI #Sentra #TogetherFund #a16z #ContextGraphs

71

3 Comments

Excited to share that Pinterest has entered into an agreement to acquire tvScientific, a leading connected TV (CTV) performance advertising platform. This acquisition will build on our existing partnership with tvScientific and expand how Pinterest can help advertisers reach people across multiple screens, including connected TV. By combining Pinterest’s AI-powered intent signals with tvScientific’s outcome-focused CTV technology, advertisers will be able to measure how TV drives results alongside search, social and Pinterest campaigns. This is another step forward in our mission to help people move from inspiration to action while giving brands the tools to reach them more effectively. Integrating tvScientific’s technology into our automation and AI-powered advertising products will accelerate our work to build a full-funnel, multiscreen performance solution. I’m thrilled about the possibilities ahead and what we’ll build together with tvScientific. Read more here: https://lnkd.in/gXcqFTU7

1,416

22 Comments