Responsible Generative AI
There are several potential uses of AI to improve our productivity , better the decision-making process , improve health care , accelerate literacy , enhance communication and so on .
Besides protecting the power of GenAI from hallucination , bias , improper usages , criminal activities , and other harmful impacts on human society and environment , it is important to promote “ responsible ” and “ helpful ” use cases . Some of those are already mentioned in the previous sections . However , it would be appropriate to give some additional focus on “ desirable ” applications of GenAI .
These can be in many diverse areas with the common threads of :
• Automating mundane tasks which would improve the productivity of people who can devote the time and effort for better tasks .
• Extending the human cognitive capabilities for complex tasks such as new discoveries , closer monitoring , better diagnosis , better communication , enhanced creativity , etc .
• Enabling services that are hitherto not available for people with access deficiencies .
• Enabling services which are not reachable to parts of the world because of affordability , lack of education or awareness .
• Enabling better interfaces to tools , appliances , machines , and applications .
Many use case examples can be found in fields of education , scientific research , medicine , communication , art , manufacturing , transportation , and so on . Leaving the hype and fear mongering aside , as “ responsible technologists ,” let ’ s all work together to determine how this powerful new tool can be “ tamed ” for the benefit of humanity in the coming years .
REFERENCES
[ 1 ] Yun-Cheng Wang , “ An Overview on Generative AI at scale with Edge Computing ,” July 2023 . https :// arxiv . org / pdf / 2306.17170 . pdf
[ 2 ] " What are AI hallucinations ?, 2023 . https :// www . ibm . com / topics / ai-hallucinations
[ 3 ] P . Park , et . al , “ AI Deception : A Survey of Examples , Risks , and Solutions ,” 2023 . https :// arxiv . org / pdf / 2308.14752 . pdf
[ 4 ] OpenAI , " GPT-4 Technical Report ", OpenAI , 2023 . https :// arxiv . org / pdf / 2303.08774 . pdf
[ 5 ] M . Turpin , et . al , " Language models don ’ t always say what they think : Unfaithful explanations in chain-of-thought prompting .", arXiv : 2305.04388 [ cs . CL ], 2023 . https :// arxiv . org / abs / 2305.04388
38 March 2024