🌈Alignment Bears Care About the Future of AI 🌈
5050 AI Align-A-Lot Program helps AI safety researchers and bear-gineers build startups for a caring and aligned future.
Psst! Visit our non-carebear version here.
Applications are open until Sep 20. US and UK cohorts run end of Sept - Dec 2025.
Recommend someone to 5050 AI Align-A-Lot here.
We believe the most powerful things in the world should also be the most caring. But right now, AI development is racing ahead faster than a Runaway Bear cloud car, and safety research is struggling to keep up.
We're here to change that, one caring researcher at a time.
What If We Could Make AI Safety As Irresistible As a Bear Stare? ✨
Care Bears taught us that caring is powerful. Instead of fighting against the forces that drive technology forward, we're going to make safety so valuable, so profitable, and so essential that everyone will want it.
Because the biggest problems create the biggest opportunities to help.
Your Safety Research Could Become the Next Friendship-Powered Company 💕
You've spent years learning how to make AI systems caring, trustworthy, and aligned with human values. But turning that beautiful research into a company that can spread caring throughout the AI world?
That's where your Alignment Bear friends come in.
The Proven 5050 Magic, Now With Extra Alignment Caring Power 🌟
5050 is the world's best rainbow bridge for scientists and engineers to start deep tech startups. For years, we’ve helped scientists and engineers turn their research into companies that make the world better. Our alumni have built incredible things—from super-fast DNA sequencing (like Tenderheart Bear's caring for health!) to carbon-negative factories (like Friend Bear's love for the planet!).
340+ caring researchers have joined our 5050-a-Lot community. 78 companies have launched with friendship power. Many wouldn't exist without the caring support of 5050.
Now we're bringing all that caring magic to the most important mission of our time: building companies that keep AI safe and friendly.
Top AI Alignment Bear Mentors 🤖
Wojciech Zaremba,
co-founder of OpenAI
Brave Heart Bear. Co-founded OpenAI and led development of GPT-1, GPT-2, teaching AI to communicate with care. Previously at Facebook AI Research and Google Brain. PhD focused on deep learning from NYU under Yann LeCun.
Jaan Tallinn,
co-founder of Future of Life Institute
Guardian Bear & Protector of Care-a-Lot's Future. Founding engineer of Skype and Kazaa (teaching computers to share friendship!). He also co-founded the Centre for the Study of Existential Risk at Cambridge University, and philanthropically supports organizations protecting Care-a-Lot from scary No-Heart-level threats.
Emmett Shear,
co-founder of Softmax
True Heart Bear. Former interim CEO of OpenAI and co-founder of Twitch (acquired by Amazon for $970M). Now co-founder of Softmax, teaching AI systems to align with human values and care about what we care about!
Bryan Johnson,
founder of Don’t Die
Forever Friend Bear. Founder of Braintree Venmo (acquired by PayPal for $800M worth of caring). Founder of Blueprint, helping everyone live long enough to see AI become truly friendly. Most measured person in history. Goes to bed on time (because self-care matters!).
Ross Girshick,
co-founder of Vercept
Vision Bear. The 18th most cited Bear in the history of science. Deep learning pioneer who taught AI to see with care (Faster R-CNN, Mask R-CNN, Segment Anything). Previously Research Scientist at Meta AI and AI2, spreading visual friendship everywhere.
Ronnie Chatterji,
Chief Economist at OpenAI
Prosperity Bear. Led the $52B CHIPS rollout to bring caring technology to everyone. Served as Biden's Deputy Director of the National Economic Council and Chief Economist at the Department of Commerce. Professor of business and public policy at Duke, teaching that caring is good business.
Dileep George,
co-founder of Vicarious
Smart Heart Bear. Co-founded Vicarious ($250M raised, acquired by Alphabet) to build AI that thinks with both logic and love. Co-founder of Numenta & pioneer in neuroscience-inspired AI that cares. PhD from Stanford on teaching machines hierarchical friendship.
Boaz Barak
Harvard Prof and Member of the Technical Staff at OpenAI
Proof Bear. Harvard Prof & Member of Technical Staff at OpenAI, Professor of CS at Harvard, proving mathematically that AI can be caring, safe, & powerful. At OpenAI, he works on AI alignment and safety with the power of friendship theorems. His research ensures AI foundations are built on caring principles.
Kanjun Qiu
co-founder of Imbue
Empowerment Bear. She's teaching AI agents to reason and code with the power of human understanding! Built Imbue to create AI that truly collaborates with people (raised $200M in caring capital). Partner at Outset Capital, nurturing the next generation of safety-focused AI startups. Her current focus is empowering humans to stay in control as AI becomes more powerful.
Areas We Want to Support in Align-a-Lot: 🌟
Safety Testing & Care-Checking
Tools that give AI systems regular "Care Bear Stare" check-ups for alignment, robustness, and safety before and after deployment.
Interpretability Rainbows
Software that makes AI decision-making as transparent as a rainbow, essential for healthcare, finance, and other areas where caring matters most.
Alignment Infrastructure
Developer tools that make it easier to build AI systems that naturally care about human values - like built-in Care Bear powers!
Governance & Caring Compliance
Systems that help organizations deploy AI with love and responsibility while meeting emerging caring requirements across all sectors.
Human-AI Friendship
Interfaces and workflows that keep humans and AI working together like best friends, especially for important decisions.
Robustness & Protection Shields
Solutions that give AI systems Care Bear-level protection against attacks or confusion, keeping them reliable and caring.
Bio-Safety & Caring Science
Platforms that ensure AI helps with biological research in caring ways, preventing misuse while enabling healing breakthroughs.
AI-Powered Care Defense
Defense systems that protect against mean AI attacks, secure caring AI deployment, and prevent No-Heart-style manipulation.
And more caring innovations!
Who Should Join 5050 AI Align-a-Lot? 💖
You'd be perfect for AI Safety 5050 Align-a-Lot if you:
Have expertise in making AI more aligned, interpretable, robust, or caring
Believe your research could solve real safety problems in the world
Want to build a company that spreads caring throughout the AI ecosystem
Are curious about starting a company but need friends to guide you
Have an idea for a safety-focused startup (or want to develop one with us!)
Share our belief that the most powerful AI should also be the most caring
Whether you're validating an idea or ready to build — Align-A-Lot is for you! 💕
You'll work closely with the Alignment Bears team to answer:
How do I turn AI alignment research into a friendship-powered startup?
How do I align commercial success with a caring future?
How do I recruit a world-class team of caring builders?
Am I the right Care Bear to do it?
You'll explore if entrepreneurship is your true calling, what it takes to be a caring founder, and how to develop the traits all great Alignment Bears share. You'll learn how to translate your research into a startup that spreads caring, build with love and speed, de-risk early with friendship, and avoid No-Heart's costly mistakes.
You'll build alongside a cohort of ambitious, caring scientists and engineers. Through our curriculum designed for technical Care Bears, weekly rainbow workshops, access to top AI mentors, one-on-one mentorship from Alignment Bear Partners, visits to successful caring startups, and a 3-day Care-a-Lot camp, we'll get you ready to start a safety-first startup.
Align-A-Lot is entirely free. No equity, no fees, no Dark Heart clauses.
Our next cohort runs from Sep/Oct to Dec '25. Sessions are held in SF (Rainbow Francisco), Boston (Care-ston), London, and remotely from anywhere in Care-a-Lot.
The Align-A-Lot Adventure
14 Weeks of Caring Entrepreneurship 🌈
1. Identify civilization-scale problems that spread caring
Align-A-Lot will help you answer: How do I turn AI alignment research into a caring startup? How do I make safety profitable AND friendship-powered? We'll help you identify problems you can uniquely solve with love.
Key questions we'll explore:
Does your idea create natural caring milestones?
Is it caring enough, or TOO caring?
Is there a market for responsible AI that cares?
Do the care-conomics make sense?
Is your solution 10x more caring, safer, or friendlier than current approaches?
2. Learn everything you need to start a caring AI safety startup
Weekly entrepreneurship workshops in the Cloud Castle. You'll learn the best approaches to:
Hiring team members who care
Fundraising with friendship
Engaging customers who value caring
Executing with both speed AND care
Storytelling that spreads the caring mission
Alignment Bear Partners will meet with you weekly for mentorship, resolving challenges with Care Bear Stares, and setting the pace for caring execution.
3. Grow into a Great Caring Founder
We'll help you figure out if entrepreneurship aligns with your caring nature (it's not for every Bear!) and coach you through the early days. You'll meet successful, caring founders and learn what building with friendship looks like.
Personal growth workshops will help you develop traits all great Alignment Bears share like thinking with both logic and love, setting high caring standards, moving fast without losing compassion, persistence with purpose, making work a caring mission, and resourceful friendship.
4. Join a Community of Future Alignment Bears
You'll join a cohort of ambitious, caring scientists and builders tackling AI's biggest safety challenges. Building alongside other Alignment Bears will challenge you to level up your caring while providing a support network of true friends.
Since you're all navigating the caring entrepreneurship journey together, someone in your cohort has likely just solved the exact challenge you're facing - with friendship!
5. What? When? Where?
Next cohort: September to December 2025
Kickoff weekend: Late September in Rainbow Francisco, Early October in London
3-day Care-a-Lot retreat: October 24-26 in Rainbow Francisco, November 14-16 in London
Weekly sessions: Office hours with Alignment Bear Partners and afternoon workshops
Ready to turn your AI safety research into a caring, world-changing company? Join the Alignment Bears!
Apply now at https://www.fiftyyears.com/5050/ai
Applications are open until September 20. US and UK cohorts run end of Sept - Dec 2025.
Recommend someone to 5050 AI Align-A-Lot: https://50y.typeform.com/ai-recommend
Questions? Send a caring message to ale@50y.com.