AI ethics for tech writers
summary
Claims that generative AI can cut support costs tenfold and auto-generate docs are paired with concrete tools like ChatGPT DALL·E Bing and a real-world failure where Mozilla pauses its AI help; outputs often hallucinate or plagiarize, so mitigation requires human review and institutional checks rather than handwaving. Training commonly scrapes internet data without consent and one LLM’s training emits CO2 comparable to 150 round trips Beijing–New York, raising systemic harms tied to accountability displacement and the need for model card transparency.
AI-written summary · about
chapters (9)
- 0:00AI ethics in tech writing
- 2:10Generative AI and ethics
- 7:10Generative AI for docs
- 10:22Accountability in AI tools
- 12:32AI tools for users
- 16:31Ethics on communities
- 19:45AI energy impact
- 22:16Ethical considerations
- 24:29Q&A
related talks
28mEthical AI?!
43mAI Engineer Ethics: Shaping Responsible Generative AI
56mPANEL: Unleashing the Future: The Impact of Gen AI on Development Productivity
31mGenAI among us
38mBuilding Responsible AI in the Age of GPT
13mMeet the Speakers
28mIs GenAI the technology revolution we need? How can we harness it at work?
38mThe Cognitive Cost of AI-Assisted Open Source