Disruptive Innovations comes with greater responsibility as it brings high risk before it matures...!!
ChatGPT - An AI enabled Chatbot which went viral within hours of its launch & is capable of disrupting some industries in coming months.
Its built on top of OpenAI’s “GPT-3” family of large language models & is fine-tuned with both supervised and reinforcement learning techniques.
Its a great innovation, for sure slated to disrupt few industries in a long run. Sharing my views based on some research I did on its capability, current shape & future possibilities.
Advantage
It has garnered worldwide attention for its efficient and precise ability to write emails, essays, poetry and even generate lines of code based on a prompt.
Gives very specific generic response to the questions or queries based on the “Prompts” as it’s a data trained model.
Industries at risk of disruption
• Search& AdTech, Suggestive Coding&Hacks
• Education, EdTech, HR & Recruitment etc
• Role Plays - Storyteller, Tech Reviewer, Profile Builder etc
• Content Creation, MarTech& Branding
• Social Media Influencer & Communicator
• Customer Services - Conversation AI driven Bots
Current Limitations:
• Very initial raw version, yet to mature. Microsoft is focussed to make it big & is pumping money to get it to next level.
• Can’t give answers to latest questions as is trained on data till 2021.
• Response is limited to generic information, not very analytical & tactical direction. It's not perfect but it is a great starting point to make life easier for few things.
• Data accuracy & integrity is still questionable.
• Security & data privacy is a big concern.
Inference:
Though many organization & startups are jumping on to it, to try it out for one or the other possible business use cases, in my views, it still has little long way to mature as a robust plug & play model.
It will for sure have major impact on customer services using chatbots which is currently having lot of limitations, with generative AI driven conversational bots like ChatGPT, it will prove to be more convenient & productive for the organizations as well as improve customer experience & engagement.
Using such models, will reduce the manual interventions, make interactions more meaningful, engaging & efficient with higher efficiency with lesser manual interventions, lower customer support. Thesupport manpower will be able to focus more on strategies leading to better conversions, brand consistency & brand loyalty.
Though there are more than 20 odd AI driven chat bots already in market competing with each other’s, with ChatGPT coming in, there will be even bigger push in this segment. "GoogleBard" - backed by Google could be an immediate competition/threat, which though wasn't able to make the equally viral buzz to meet the high expectations as it was a hush-hush launch but google being the index of internet (worldwide content) is not going to let it go easily, it will have a big push to GoogleBard to give more accurate, specific, analytical response.
There has been big concerns over its security, data accuracy & integrity as it’s a trained model against a data set which can go wrong & this has been seen in many cases. As its based on open source technology, there is higher vulnerability & risk of personal information leakage which has recently found & admit by OpenAI.
As more and more startups innovate and get into market faster there is higher threat of breach. It should be ideally released in a controlled environment, ensure that its secure & robust, before getting out to whole world.This needs better regulations to restrict the abuse technologies by bad players.
"Many universities & countries are banning ChatGPT as it will impact the learning ability & skill development of students as aiding to easy way out of capturing information in public space & capable of doing activities like writing code, writing essays etc impacting the creativity in new generation."
Also it is trained on public data so its authenticity of enterprise business data & information is very questionable.
Also it has been observed that many universities & countries are banning ChatGPT as it will impact the learning ability & skill development of students as aiding to easy way out of capturing information in public space & capable of doing activities like writing code, writing essays etc impacting the creativity in new generation.
But in very diverse complex demography like India with culturally & regional diversity with multiple lingual it is hard problem to solve as these models would be primarily English dominance. AI bias & lack of empathy is another challenge which is difficult to crack.
Having said this, in coming years, it will shape up right to address these limitations & concerns,bring a balancing act between the technology innovation putting human intelligence & empathy at the centre to address diversified complex needs.
It will for sure the current job paradigm in some industries & would require reskilling& focussing on more productive & value add rather than doing job which could be automated with ML & AI.
As any such technology innovation brings disruption, comes with lot of responsibility as brings higher risks of cyber threat & data breach before it matures with sufficient governance & controls.
Its built on top of OpenAI’s “GPT-3” family of large language models & is fine-tuned with both supervised and reinforcement learning techniques.
Its a great innovation, for sure slated to disrupt few industries in a long run. Sharing my views based on some research I did on its capability, current shape & future possibilities.
Advantage
It has garnered worldwide attention for its efficient and precise ability to write emails, essays, poetry and even generate lines of code based on a prompt.
Gives very specific generic response to the questions or queries based on the “Prompts” as it’s a data trained model.
Industries at risk of disruption
• Search& AdTech, Suggestive Coding&Hacks
• Education, EdTech, HR & Recruitment etc
• Role Plays - Storyteller, Tech Reviewer, Profile Builder etc
• Content Creation, MarTech& Branding
• Social Media Influencer & Communicator
• Customer Services - Conversation AI driven Bots
Current Limitations:
• Very initial raw version, yet to mature. Microsoft is focussed to make it big & is pumping money to get it to next level.
• Can’t give answers to latest questions as is trained on data till 2021.
• Response is limited to generic information, not very analytical & tactical direction. It's not perfect but it is a great starting point to make life easier for few things.
• Data accuracy & integrity is still questionable.
• Security & data privacy is a big concern.
Inference:
Though many organization & startups are jumping on to it, to try it out for one or the other possible business use cases, in my views, it still has little long way to mature as a robust plug & play model.
It will for sure have major impact on customer services using chatbots which is currently having lot of limitations, with generative AI driven conversational bots like ChatGPT, it will prove to be more convenient & productive for the organizations as well as improve customer experience & engagement.
Using such models, will reduce the manual interventions, make interactions more meaningful, engaging & efficient with higher efficiency with lesser manual interventions, lower customer support. Thesupport manpower will be able to focus more on strategies leading to better conversions, brand consistency & brand loyalty.
AI driven conversational bots like ChatGPT, it will prove to be more convenient & productive for the organizations as well as improve customer experience & engagement.
Though there are more than 20 odd AI driven chat bots already in market competing with each other’s, with ChatGPT coming in, there will be even bigger push in this segment. "GoogleBard" - backed by Google could be an immediate competition/threat, which though wasn't able to make the equally viral buzz to meet the high expectations as it was a hush-hush launch but google being the index of internet (worldwide content) is not going to let it go easily, it will have a big push to GoogleBard to give more accurate, specific, analytical response.
There has been big concerns over its security, data accuracy & integrity as it’s a trained model against a data set which can go wrong & this has been seen in many cases. As its based on open source technology, there is higher vulnerability & risk of personal information leakage which has recently found & admit by OpenAI.
As more and more startups innovate and get into market faster there is higher threat of breach. It should be ideally released in a controlled environment, ensure that its secure & robust, before getting out to whole world.This needs better regulations to restrict the abuse technologies by bad players.
"Many universities & countries are banning ChatGPT as it will impact the learning ability & skill development of students as aiding to easy way out of capturing information in public space & capable of doing activities like writing code, writing essays etc impacting the creativity in new generation."
Also it is trained on public data so its authenticity of enterprise business data & information is very questionable.
Also it has been observed that many universities & countries are banning ChatGPT as it will impact the learning ability & skill development of students as aiding to easy way out of capturing information in public space & capable of doing activities like writing code, writing essays etc impacting the creativity in new generation.
But in very diverse complex demography like India with culturally & regional diversity with multiple lingual it is hard problem to solve as these models would be primarily English dominance. AI bias & lack of empathy is another challenge which is difficult to crack.
Having said this, in coming years, it will shape up right to address these limitations & concerns,bring a balancing act between the technology innovation putting human intelligence & empathy at the centre to address diversified complex needs.
It will for sure the current job paradigm in some industries & would require reskilling& focussing on more productive & value add rather than doing job which could be automated with ML & AI.
As any such technology innovation brings disruption, comes with lot of responsibility as brings higher risks of cyber threat & data breach before it matures with sufficient governance & controls.
