lablab.ai

LabLab is bringing AI Hacks to Web3. We are the largest global hackathon platform with over 250k+ builders, and foundational partnerships with @Google, @Nvidia, @Meta and many more. We’ve proven our tracks; partnering with @Circle to co-launch @arc, shipping 4k +new apps in two weeks. It’s time to build with the best.

One quarter. $300 billion in VC money. 4 of the 5 largest venture rounds in history - all Q1 2026. OpenAI $122B. Anthropic $30B. xAI $20B. Waymo $16B. The capital is concentrating fast. Full breakdown → lablab.ai/ai-articles/la….

87% of enterprise workers are avoiding AI tools. 54% bypassed company AI and did the work manually. 33% haven't touched AI at all. Companies bought the tools. Nobody is using them. That is not an AI problem. That is a trust problem.

The AI world just changed. Again. Here is everything that happened this week. GPT-6 imminent. Anthropic's wildest 72 hours ever. 87% of enterprise workers avoiding AI tools. $300B in VC money. Deep dive →

Stop building in silence. Build in public. Ship. Show proof. Your product is the application. Your traction is the pitch. @Surgexyz_ Ignition is live. LabLab Builders - that's your call. $250,000 on the table. Join the race -> surge.xyz/challenge

Excited to share PennyPilot, my project for the Nano Payments x Arc hackathon with @buildoncircle, @arc, and @lablabai PennyPilot explores a simple but important problem: AI agents are becoming capable of making useful economic decisions, but giving an agent direct wallet access is not enough. In real finance workflows, every automated spend needs limits, context, accountability, and a receipt trail that can be reviewed later. I am built PennyPilot as a spend-control card for AI finance agents. A user defines a task policy first: total budget, maximum price per paid fact, approved providers, and the number of facts the agent is allowed to buy. The agent can then purchase small financial facts with USDC through Circle Gateway x402, while staying inside the policy set by the user. A key part of the product is batch authorization. Instead of asking the user to sign every proof one by one, PennyPilot lets the selected batch share one authorization. That makes the experience much more realistic for agent workflows where dozens of low-cost API calls may be needed, but manual approval for each one would destroy the UX. On the verification side, provider receipts are anchored on Arc. This gives the user and reviewer a clear audit trail: what was bought, from which provider, under which policy, and at what cost. The goal is not just to move money, but to make machine spending observable and controlled. For me, PennyPilot is a step toward supervised machine spending: useful agent autonomy, strict payment guardrails and verifiable receipts for the next generation of agentic commerce.

More news coming soon. Join the challenge → lablab.ai/ai-hackathons/….

Best use of Gemini now wins: San Francisco Builders: $2,000 · $1,400 · $800 Online Builders: $1,200 · $600 Agent thinks. Agent transacts. Agent pays. Sub-cent. Sub-second. No gas.

Agentic Economy on @arc Hackathon: Day 1. Watch @GoogleDeepMind dropping extra $5K into the Prize Pool. Power up apps built with @BuildOnCircle Nanopayments with @GeminiApp & GoogleAIStudio. Settle value on Arc in USDC. Let’s turn up the heat.

Getting Started on AMD Developer Cloud | AMD Workshop x.com/i/broadcasts/1…

Automating the Software Development Lifecycle with AI x.com/i/broadcasts/1…

Agents shouldn’t be rate-limited. @AIsaOneHQ → 100+ endpoints @BuildOnCircle → per-call payments @arc → instant settlement That’s the stack. Now go try it ↓ lablab.ai/ai-hackathons/….

More news coming soon. Join the challenge → lablab.ai/ai-hackathons/….

The full stack is coming together: @arc settlement. @buildonCircle Nanopayments. @aimlapi intelligence across text, image & multimodal. Agent thinks. Agent transacts. Agent pays. Sub-cent. Sub-second. No gas.

Agentic Economy on @arc Hackathon: Day 1 just keeps giving. @aimlapi is dropping $10 in free credits for every participant. One API. Hundreds of models. Text, image, multimodal. Build agents that pay per call - not per subscription.

Stack it: @FeatherlessAI → reasoning @BuildOnCircle Nanopayments → value @arc → execution And build → lablab.ai/ai-hackathons/….

That’s the real upgrade. Not “better prompts”. A routing layer across thousands of models optimized for cost, latency, and task. No infra to manage. No setup overhead.

Most agents are still stuck on one brain. Now @FeatherlessAI joins the Agentic Economy on @arc hackathon. Unlocking access to specialized models across coding, medical, legal, multilingual. $25 credits for 1,000 builders. Extra 900 inference credits on the table. To let you build agents that adapt per task - not just guess.