Responsible by Design
summary
Names a concrete concentration: 94% controlled by five companies, and pinpoints three product failures — ethics washing, opaque models and technical gatekeeping. Describes BigScience BLOOM as a 176B-parameter model built by 1,000+ researchers across 70 countries with full training corpus docs, a responsible AI license banning surveillance and disinformation, and multilingual data for 46 languages. Proposes practical fixes: local-by-default deployment, royalties for data contributors, and community governance.
AI-written summary · about
chapters (9)
- 0:00Introduction to ethical AI
- 1:30AI ethics as power question
- 3:20Problems with current AI ethics
- 4:50Free software principles for AI
- 6:38BigScience Bloom alternative
- 8:26Three provocations for ethical AI
- 10:11Structural ethics over individual responsibility
- 11:16Concrete actions for change
- 12:24Key takeaways on ethical AI
related talks
40mSocially responsible AI: Practical tools for robust design
49mPanel Discussion: Navigating the Ethical Landscape: Responsible...
34mResponsible AI: The Key to Ethical, Safe and Inclusive Software Development
01h 16mLifting the Hood on Big AI - The Future of Transparency and Accountability in AI
35mFrom hype to responsibility: what works & matters for whom in data and AI?
30mResponsible AI – Principles & Practice
18mDeveloping AI in a privacy conscious world
29mResponsible AI Challenges and Recommendations