The General Partner model and the startup model are not the same game. Startups are supposed to chase huge, ambitious targets, knowing that most who try will fail. General Partners run diversified portfolios specifically so they don't lose their liquidity providers' money. Many founders grind themselves down trying to get their investors some kind of return when the investors have already written the company off a long time ago. These founders are better off shutting it down and figuring out the next thing.
The defensibility question can become a trap at seed. Most great companies do not start with a perfect moat. They build one by learning faster, hiring better, earning trust, and executing for longer than everyone else. The founder takeaway is not to ignore moats. It is to know when “what stops OpenAI from building this?” is a real risk, and when it is just a lazy way to underrate speed, taste, and customer obsession.
YC CEO Garry Tan reframes what Y Combinator actually delivers by borrowing psychologist Alice Miller's concept of the "enlightened witness," the single strongest predictor of whether traumatized children grow into functional adults: not someone who could fix or rescue them, but someone credible enough to confirm their reality.
Unusual Ventures Director of Programs & Community Niamh O'Donnell shares the seven-step playbook from arriving in San Francisco at 24 on a one-year Irish visa with no tech network to hosting rooms full of founders, investors, and Fortune 500 leaders five years later.
Researchers at Kellogg tested an alternative social media algorithm on 2,000 Bluesky users around the 2024 US election and found it reduced toxic and politically extreme content without moderating any specific posts. The algorithm worked by simply limiting the outsized influence of a small number of highly engaged accounts, making feeds more representative of the broader conversation on the platform. Users on engagement-based feeds saw 57.3% more political posts after the election compared to those on chronological feeds, and they significantly overestimated partisan animosity in their networks. Participants using the alternative algorithm reported enjoying the platform more, challenging the industry assumption that engagement-driven content is necessary to retain users.
2026.05.29·
insight.kellogg.northwestern.edu·5 min read·TLDR Marketing
US grocery prices are rising again as food inflation runs above average. Costs are being pushed up by extreme weather, tariffs, a shrinking US cattle herd that sent beef to record highs, and supply disruptions that also lifted items like tomatoes by 33% in recent months. The outlook will worsen into 2027 as drought spreads across major crop regions, fertilizer prices are up 20% since the Iran war, and a likely El Niño threatens global harvests. With wages softening, savings falling, and household debt rising, grocery bills are set to keep outpacing incomes.
A drop in TikTok views can happen when the algorithm lacks enough signals to categorize a video. TikTok now allows keywords to be added after posting, offering a cleaner way to guide distribution without forcing SEO into captions or visuals. Updating keywords helps the algorithm better understand content, and in some cases, has revived videos that previously had little reach.
A guide to combining media buying, creative testing, reporting, and competitive research into a single workflow inside Claude. The strategy centers on connecting live Meta ad data, automated competitor monitoring, and AI-generated creative so you can audit accounts, spot fatigue, and launch new tests without switching tools. The biggest advantage comes from systemizing context through brand-specific Claude skills, shared GitHub learnings, and scheduled research tasks so every campaign builds on past performance instead of starting from scratch. It also allows for faster creative iteration since teams can generate and refine new ad concepts in seconds, then tie those outputs directly to live campaign dashboards and performance data.
TLDR is hiring a VP of Marketing to scale growth, pipeline, and brand across 8M+ subscribers, 13 newsletters, and advertisers like AWS, Google Cloud, Microsoft, and Anthropic. Apply here [8].
The Ahrefs AI Search Benchmark Report analyzes 100M+ data points across AI Overviews, AI Mode, and ChatGPT. It covers AI citations, visibility signals, user traffic behavior, and more. The research finds that YouTube mentions correlated most strongly with AI visibility, and Google sends 190x more traffic than ChatGPT.
AI platforms overwhelmingly rely on Yelp when answering local business queries, citing it 3.4 times more than the next closest competitor across four major AI models, according to an analysis of 28 million responses in Q4 2025. Yelp captured 72.5% of all local discovery citations on Google AI Mode and held a 62.1% share on Perplexity. Nearly all citations stemmed from category-level queries like "best cleaning service near me" rather than brand searches.
Mystery is driving modern cultural breakout. The strongest campaigns leave gaps that audiences fill with theories, edits, and lore. Backrooms shows how unfinished narratives become shared worlds where the story expands through community participation. Breakthrough now comes from participation, not reach. People want to feel early, decode meaning, and share discoveries before they fully spread. The more brands hold back and trust audiences to connect the dots, the more culture forms around them.
Crocs CMO Terence Reilly leaned into the brand's reputation as "ugly" footwear, reframing widespread ridicule as proof of awareness that could be converted into cultural relevance. He credits frontline employees for some of the brand's biggest moves, including the Post Malone partnership, which started when an intern flagged a photo of the artist wearing Crocs. Reilly pushes teams to take risks by publicly celebrating both wins and failures, arguing that most organizations talk about failing fast but structurally punish it. His track record includes a similar turnaround at Stanley, where a single TikTok-driven campaign generated $50 million in revenue and 100 million views.
The new feature gives advertisers instant policy feedback while building ads, replacing the traditional post-submission review queue that often delays campaign launches.
Product teams are shipping AI-powered features faster than ever. But how do you experiment, release, and optimize AI experiences without increasing risk?
PMs are not “dying” but changing in scope depending on the organization. This article shares a framework for PMs' past, present, and future across several areas, with special emphasis on how AI and AGI pressure will reshape responsibilities.
AI tools are changing product development workflows. New approaches move teams from early ideas to building and production, showing what changes across the product lifecycle.
AI products often lose users not because the model is weak, but because adoption design decisions are missing or wrong. There are four common failure modes: activation cliffs, a trust gap, broken correction loops, and an agent autonomy ceiling.
Anthropic and OpenAI have achieved PMF for enterprise coding/general-purpose agent tools. The shift is driven by real enterprise demand for products like Claude Code/Cowork and Codex, despite thin “AI cost failure” anecdotes.
Tired of babysitting agents? With OpenProse [11], you define your outcome ("Inbox at zero by 5pm") - the runtime executes, watches for drift, and pings you when it's done. Write your goal once in plain English, let the system keeps it true. To start, simply open terminal
To achieve a big long-term goal, teams must “sell Point C”- a concrete, vivid next destination they will reach in the next 12–24 months - rather than settling for “continue existing.” This should be a strategic, fundable milestone aligned with mission, vision, strategy, and anchored in the needs of the community the team serves.
User segmentation gets more design buy-in than relying on personas alone. Personas often get dismissed, and there are four segmentation elements to reinforce personas: total audience size, time period/seasonality, device/tech context, and new vs. returning customer.
2026.05.29·
thedesignersfieldguide.substack.com·7 min read·TLDR Product
Slack evolved its AI infrastructure from self-managed AWS SageMaker to managed Amazon Bedrock to solve scaling and latency issues. By moving to Bedrock, Slack reduced engineering overhead and improved efficiency through Provisioned Throughput and on-demand capacity.
AI labs will own many horizontal AI tools, but startups can still win in complex vertical workflows. The strongest AI app companies will own the system of work, including customer data, governance, integrations, and measurable business outcomes.
Cognition raised over $1B at a $26B valuation, with significant backing from major investors to expand Devin, an AI software engineer. Devin has significantly cut project times and improved automation for clients like Mercedes-Benz and Itaú. Cognition aims to further streamline software development by matching models to tasks while expanding its engineering capabilities.
Biohub has made its open discovery engine for protein structure prediction, design, and biological discovery available to researchers everywhere. The release includes ESMC, a state-of-the-art language model that has internalized the fundamental properties that govern protein biology; ESMFold2, a design engine designed to transform ESMC's sequence representations into atomically-resolved 3D structures of biomolecular complexes; and ESM Atlas, which makes ESMC's representations navigable across 6.8 billion protein sequences and 1.1 billion predicted structures. All three models are freely available to the global scientific community.
The blog post introduces a method to reduce the weight synchronization payload in async RL using "Delta Weight Sync," which transmits only changed model parameters between RL steps, significantly reducing data transfer from gigabytes to megabytes. A Hugging Face Hub "bucket" manages high-frequency object storage, enabling separate locations for the trainer and inference engine without direct communication, leading to substantial bandwidth savings.
Real-world systems often behave differently in production than they do in the lab. Teams often discover these failures after launch, then spend weeks fixing them. That feedback loop is slow and manual. Today, it is possible to build agents that self-improve. This post looks at how OpenAI used Codex to build this type of agent at Thrive Holdings, resulting in an AI that can prepare increasingly complex tax returns.
Both Anthropic and OpenAI have started aggressively pricing their APIs. This is likely because they have found product-market fit with coding/general-purpose agent products. Companies spending over $200 per month per user helps these businesses cover their costs much better than charging $10 to $20 per month per user. Coding agents amplify this spending significantly.
Learn on how to build strong data foundations for agentic AI in this executive panel[IS1] . Get strategies for databases, governance, and AI implementation from experts at Yahoo, Mercedes-Benz, Regeneron, and
Secure MCP Tunnel enables connecting private MCP servers to OpenAI products without exposing them to the internet. It uses tunnel-client to establish outbound HTTPS paths for request handling while maintaining server privacy. The solution integrates easily with existing systems, supporting enterprise networking requirements and maintaining secure data flow.
Apex is a React Native coding model trained to build apps by analyzing architecture decisions, fixing framework-specific issues, and reasoning about constraints. While it doesn't match frontier models on coding benchmarks, the optimized model significantly alters the performance-to-cost ratio within its specific domain. The model is still in development. It is now available in a private beta with selected teams.
LiteParse is a standalone OSS PDF parsing tool that provides high-quality spatial text parsing with bounding boxes without proprietary LLM features or cloud dependencies. It features fast text parsing, screenshot generation, and support for multiple languages, platforms, and output formats. Everything runs locally on users' machines.
Nvidia will invest $150 billion a year to make sure that Taiwan remains at the epicenter of the AI revolution. The investment is aimed at cementing Taiwan as the world's tech manufacturing hub for a long time. Nvidia will create a new headquarters in Taiwan to expand its partnership with TSMC, benefit from close proximity to advanced packaging technology not yet available at TSMC's factories in the US, and boost its alliances with other nearby partners. Expanding the AI ecosystem helps Nvidia further its bottom line.
YouTube said it will automatically apply labels to videos containing significant photorealistic AI content, reducing reliance on creator self-disclosure.