We are proud to partner with @NSDCINDIA to work towards improving the skilling systems and ensuring the AI readiness of @NSDCINDIA through capacity-building initiatives on the responsible use and application of AI across domains.
ETHICS
-
Understanding AI Limitations for Ethical Development
By
–
AI comes with it's own set of limitations and @LofredM reminds us that understanding these limitations is crucial for responsible AI development. Read more on how AI can be used ethically and equitably: https://
bit.ly/3LDGtnk @jibuelias @wef @OpenAI @UniofOxford @truera_ai -
ChatGPT Thinking: Linguistic Competence and Reporting Bias Interview
By
–
Does #ChatGPT “Think”? insightful interview with Anna Ivanova @neuranna
@TwimlAI on Formal vs Functional Linguistic Competence and Reporting Bias. -
LLM Alignment Challenge: Developer Pretends to be User to Avoid Character Break
By
–
This is clever. We’re at a point in LLM alignment where, to avoid ChatGPT/GPT-4 breaking character when a user pretends to be the developer, the developer must pretend to be the user.
-
How to Raise AI: Timeless Principles Beyond Technology
By
–
Someone just referenced this article I wrote in '15 about #ai https://
bit.ly/42mA3zd So much has changed in the technology, but little in how we approach it. My examples would have changed, but not the "how to raise AI" advice…almost 8 yrs later! -
Google’s Reputation Concerns and GPT-4 Limitations
By
–
I am being harsh on Google. They care a lot about their reputation (as they should.) GPT-4 doesn't give the right answers to even its own questions. It should used as Tesla FSD :D.
-
GPT-4 Lacks Mathematical Understanding Despite Theorem References
By
–
You can see that GPT-4 has no notion of mathematics. Even when it refers to the right theorem (Chinese Remainder Theorem), it does not actually use it to generate the answer.
-
Red-teaming GPT-4: Public Transparency Over Corporate Secrecy
By
–
to start, I want to say I have nothing to gain here and I don't condone anyone actually acting upon any of GPT-4's outputs however, I believe red-teaming work is important and shouldn't be conducted in the shadows of AI companies. the general public should know the capabilities
-
OpenAI Stops Using Customer Data for Model Training by Default
By
–
Seems like they did – but rather recently. https://
techcrunch.com/2023/03/01/add
ressing-criticism-openai-will-no-longer-use-customer-data-to-train-its-models-by-default/amp/
… -
NDAs and Professional Ethics in AI Industry
By
–
Good thing I don’t sign NDAs, but it’s an important point for those who do.