Direct From TED2025: Five Insights on AI for Social Impact
Inspiration and Direction from the Frontlines
TED2025 buzzed with ideas about AI’s potential to transform our world for the better. But this excitement was countered with caution. How do we develop this powerful technology so that it benefits everyone? Clearly, opinions differ.
Many conferences use on-stage panel conversations to present various viewpoints. TED is different. It relies on a carefully curated speaker lineup to present diverse perspectives using the TED Talk format, which has become the gold standard for persuasive idea-sharing.
TED prioritizes creativity, innovation, and the human potential in its content. Those of us working at the intersection of AI and social change can look to the collective insights shared by entrepreneurs, creatives, and thought-leaders at TED2025 for both inspiration and direction.
Let’s explore five key takeaways that can help us harness AI’s potential for advancing responsible and positive social impact globally — all informed by this year’s TED speakers.
At the time of writing, many of these talks are only available through TED Live (paid content). We expect they will be made widely available on TED’s YouTube channel in the coming months, but we found the content worth the price.
1. Build Ethical Guardrails Before, Not After Deployment
The rapid advancement of AI demands proactive governance rather than reactive regulation.
Technologist Tristan Harris compared the lack of AI regulation to the lack of clarity around the downsides of social media that led to a “totally preventable societal catastrophe.” He urged: “I don’t want us to make that mistake with AI, and I want us to choose differently.” (Session 4: A Beautiful Mind)
Yoshua Bengio, a Turing Award-winning AI pioneer, stressed the need for societal guardrails. “There’s huge commercial pressure to build AIs with greater and greater agency to replace human labor. But we’re not ready… a sandwich has more regulation than AI.” (Session 1: Apocalypse Now)
Investigative journalist Carole Cadwalladr warned of the profound threats to democracy posed by unchecked technological power, stressing the urgent need to acknowledge and counter the political misuse of technology. She highlighted that we are already living in an era defined by “total information collapse” and called for immediate collective action. (Session 1: Apocalypse Now)
Eric Schmidt, former Google CEO and Chairman, acknowledged that AI development can’t be stopped. “Instead of stopping agentic work, we need to find a way to establish the guardrails… You don’t want systems that are not under our control.” (Session 11: Life Is Beautiful)
2. Design AI That Reflects Diverse Cultural Values
AI systems often emerge from narrow cultural contexts but are deployed globally.
Tech policy visionary Nanjira Sambuli highlighted the importance of culturally-rooted AI, inspired by Africa’s indigenous Ubuntu philosophy that “I am because you are.” She explained: “Data governance for us is about the meaningful participation, informed consent, self determination and community ownership of data sets from which language, nature-based knowledge, and indigenous wisdom are derived.” (Session 5: The Incredibles)
Shahram Izadi, VP and GM of XR at Google, underscored the cultural implications of creating systems that interact with people on their terms rather than forcing humans to adapt to machine interfaces. “(Computers) will share your vantage point, understand your real world context, and have a natural interface that’s both simple and conversational.” (Session 3: I, Robot)
M-PESA’s Sitoyo Lopokoiyit warned against AI becoming a form of digital colonization: “We need to open it up, build the algorithms that are relevant to Africa, and not just transport the algorithms there… if we don’t do that, it will be a form of colonization, colonialism at a scale that we’ve never seen before.” (Session 5: The Incredibles)
3. Use AI to Break Down Access Barriers
AI has unprecedented potential to democratize access to services and creative tools.
Clinical psychologist Alison Darcy founded Woebot Health, an AI therapy app, to bridge critical care gaps. “It doesn’t really matter how sophisticated the treatments are that we make if people can’t access them… We built Woebot to meet people where they’re at in those moments when it’s actually hardest to reach out to another person.” (Session 7: The Parent Trap)
Filmmaker Rob Bredow noted how generative AI is transforming visual storytelling and breaking down barriers to creative expression. “Today, we’re entering a new era of technology… Google, Meta, OpenAI — all showing sophisticated video generation systems.” (Session 3: I, Robot)
Dr. David Fajgenbaum’s EveryCure initiative uses AI to find new uses for existing drugs, bringing life-saving treatments to patients with rare diseases who might otherwise be overlooked by traditional pharmaceutical research. This AI-driven repurposing of medicines is essential in democratizing healthcare solutions. (Session 9: Inside Out)
4. Focus on Human-AI Collaboration, Not Replacement
The most promising AI applications enhance human capabilities instead of attempting to replace humans entirely.
OpenAI CEO Sam Altman acknowledged artists’ concerns about AI mimicry but asserted: “Every time… we have put better and more powerful technology in the hands of creators, I think we collectively get better creative output, and people do just more amazing stuff.” (Session 11: Life Is Beautiful)
Robotics innovator Bernt Børnich envisioned AI-driven robots that enhance human potential, creating abundant labor that would free humans from drudgery. “We are standing at the gates of a future where the work needed to build the products we use, the services we rely on and even the chores in our homes will be as effortlessly accessible as energy is today.” (Session 3: I, Robot)
Film director Jason Zada illustrated how generative AI swiftly created immersive film narratives from basic text prompts, accelerating months of work into days. He contends: “AI doesn’t replace storytellers. It gives us super-powers.”(Session 1: Apocalypse Now)
5. Embrace AI for Focused, Measurable Impact
AI applications can help social impact leaders solve specific problems with clear metrics and well-defined outcomes.
Guidewheel CEO Lauren Dunford demonstrated how AI applied to factory efficiency might seem mundane but delivers measurable climate impact. Her ‘smartwatch for factories’ approach helps optimize the millions of machines that are already running. (Session 8: Vertigo)
Inventor Boyan Slat showcased how AI is empowering the fight against ocean plastic pollution. His team deployed AI-enabled camera systems on bridges over dozens of waterways to measure trash. Feeding this data into a global model, the AI revealed that just 1% of rivers contribute about 80% of ocean plastic, enabling him to focus on the worst polluters for maximum impact. (Session 4: A Beautiful Mind)
AI is already everywhere, all at once, all the time — and it’s increasingly hard to keep up with advancements. The TED2025 speakers reminded us that the path to positive AI impact isn’t paved with algorithms alone, but with human values, diverse perspectives, and a clear understanding of real needs.
But the experts don’t agree — on a lot of things AI. This tension between “AI is amazing” and “AI is scary” was the crux of many TED2025 talks. The reality is we won’t find all the answers at any conference. But the conversations we have — and the insights we share — should help guide the decisions we make as a society.
StartingUpGood supports fresh entrepreneurial approaches to social impact. FOLLOW US on social media:
Medium | LinkedIn | X | Bluesky Social
Check out SDGCounting for the latest news on tracking the progress of the Sustainable Development Goals. #SDGs #GlobalGoals.
Note: Generative AI tools were used in the creation of this article to assist with research, summarization, and editing.
