Generative Ai: Challenges and Way Ahead
The Internet revolution of the 90s and the mobile technology boom of 2010 have brought breakthroughs in professions. The latest developments in generative AI is far bigger and will impact many businesses. AI engineers will be professionals who will use AI and machine learning techniques to develop systems that will solve problems and automate time-consuming tasks.
The second wave of transformative AI will completely change how we communicate across society. Generative AI is an AI tool that generates or creates content that can be an image, informative text, or code. This can disrupt the market as we are inputting a prompt, and this AI will output based on his learning from the data set. Data Centres Control and Store the data. Let’s say a facial recognition program. You need to train it on a number of data points, as the more data it gets trained on, the more accurate it will be. Ai learns from large data and assimilates this in a large language model. It doesn’t have the knowledge, but it has the ability to process language really fast, make a deeper association in data and precisely answer any question.
- AI tools many times generate wrong information. It’s an open research question how to fix this problem? These language models are capable of making next-word predictions, so sometimes they are not accurate, for that, you can pull in a factual context as a prompt that you can give to the model. This is how you can make it to produce more accurate information.
- Prototyping, fine-tuning, testing, evaluation, and differentiation are how all AI applications will be built in the future. In Prototyping data scientists work towards building base models through data loading, cleansing, preparation, feature engineering, model training, tuning and scoring. Fine-tuning is finding examples of output that you want and then doing a little bit of training on this base model for specialization for this task and then further fine-tuning round of this model using reinforcement learning from human feedback.
- We must steer this technology in a more ethical direction. As its capabilities increases, We can face an existential threat. There are short-term consequences where we will see social disruptions. These tools can be used for spreading misinformation. Miscreants can use it to misinform people and try to manipulate them. What anyone can do with this AI tool is make millions and billions of pieces of misinformation a day. It will be difficult for anyone to distinguish between fake and authentic information because it will provide references and sources to make it look real. So a lot of people will be be-fooled. We need to have a robust mechanism to deal with these situations.
- AI-generated deep fake videos have been a deep concern for the government and national security. There were around 15000 deep fake videos in circulation in September 2019 that have been spreading misinformation in the commoners These videos have mapped the faces of celebrities film stars sports personalities with someone else. Planted Fake videos will spread misinformation amongst the general public to defame, slander and asperse to build a narrative and perception against them. Deepfake technology is being weaponised against women. Social media firms Facebook and Youtube have been widely used by the anti-social elements to spread misinformation among the commoners.
XIM University (formerly known as Xavier University), Bhubaneswar Orissa is well known institution for providing Engineering, Management and Humanities courses. XIM University, Bhubaneswar got direct campus placements in top companies like Infosys, IBM, Accenture, Axis Bank, Cognizant and more.
Use Cases of AI
- Customer service: – Voice assistance and chat assistance can be made much as customer care executive with new-age generative AI technologies. Testings and training are going on in the labs of giant chatbot techs to make them much more finer.
- Healthcare:- The analysis of reports will be vital in cancer for providing accurate reports. Now doctor takes time to analyse those reports in hours which will be possible to do with AI in seconds. IBM Watson Health and Google DeepMind Health have been developing a large language model for the medical community that will help in businesses come up with products and services.
- E-commerce:- E-commerce platforms can make product recommendations based on user behaviour. Sales can be boosted by inducting AI in sales. Amazon has been utilizing AI technologies in Amazon-go stores and people prefer going to these stores. AI-generated deep fake videos have been a deep concern for the government and national security. There were around 15000 deep fake videos in circulation in September 2019 that have been spreading misinformation in the commoners These videos have mapped the faces of celebrities film stars sports personalities with someone else. Planted Fake videos will spread misinformation amongst the general public to defame, slander and asperse to build a narrative and perception against them. Deepfake technology is being weaponised against women. Social media firms Facebook and Youtube have been widely used by the anti-social elements to spread misinformation among the commoners. The AI market has been expanding since ChatGPT. If the strong AI market expands then there will be far much more unemployment in all sectors and then ninety percent of the jobs will get replaced by robots.
Jobs That Will Be in Threat
- SEO Job will go away: The SEO Jobs will easily be replaced by robots. The search engine has job is to refine and serve the best results to the user. After Google updates the functionality of the search engine will change. It was in a nascent stage in the early 2000s. when SEO executives were just doing the link building and just on the basis of some classified and directory links they were ranking at the top of the SERP.But now it will not work that way. You need to work in a technical segment of the website also for getting website ranked like webvitals such as Largest Contentful Paint (LCP), First Input Delay (FID) and Cumulative Layout Shift (CLS). This will even get more refined when Google brings more updates. Mostly these changes will provide a better user experience but they are usually done by the developers.
- Content Writers will have no jobs: Content Writers’ jobs are not so many in the market now as they were before chatbots. Even there are far so many content writing tools available in the market. These tools even write better than content writers. Now there are just 2-3 persons in content writing where there were 15 -20 persons even those were not well paid.
- Psudo-Developers will not have job: Now there are just 8-9 persons in development where there were 15 -20 persons This will not influence the developers much because most of the developers will make a switch to AI development.