I recently started brainstorming with my brother about what fields are actually likely to be disrupted by AI technologies that exist today and the list was quit

I recently started brainstorming with my brother about what fields are actually likely to be disrupted by AI technologies that exist today and the list was quite surprising and unlike what I had seen in the media. The impact is likely to start in the creative industries. Entrepreneurs feel free to steal : 1. Hollywood in the Cloud : The progress of computer vision algorithms, game engines like Unreal and massive computation in the cloud mean that in 25 years time you maybe able to produce a Hollywood quality movie by writing code. Unreal engine will render the backgrounds, Deep neural nets will generate the actors voices and faces and code will be used to stitch everything together. This may also include the production of background scores by neural nets primed on music in similar scenes. The number of people needed to produce a film will be cut by 10x and we will see an explosion in film making. Tik Tok is an early example of this. 2. Digital Models: This is connected to the above. You will also see digital models being used on billboards and news readers will be replaced by models like GPT-3 that convert data into narratives and then they are read by digital newsreaders. They may even make it interactive by reading out the most popular tweets or having fake discussion between AI models with different personalities. 3.Lawyers : GPT-3 has given me a lot of confidence in predicting a major disruption to legal research. You can probably semi automate case research and you don't need armies of junior lawyers or para legals to fight cases. 4.Accountants: This relies on the continued improvement in computer vision in the ability to read and interpret printed invoices. More and more transactions will happen via APIs and be shepherded by digital accountants too. 5.Programmers: I am less sure of how programmers will be replaced but there are some obvious avenues. Natural language interfaces could make most front end work obsolete. You don't really need an Uber app if GPT-3 on steroids can understand exactly what you want and then produce a widget on the fly that shows you the appropriate information on demand. Most simple apps will be folded into natural language assistants which means that front end work will go down. What does exist will be designed with the assistance of AI tools. The backend work could also increasingly be subsumed into making a knowledge base that can learn and respond to intelligent queries. 6.Therapists: Smarter NLP models could act as digital therapists. People maybe more comfortable talking to a digital therapist and not be judged by an actual person. They can be given digital bodies and voices to make them more realistic. GPT-3 is way ahead of ELIZA and even in the original ELIZA studies people became quite attached to it. Technology is making people lonely and people may turn to technology to fix it. 7.Fake twitch streamers / Cam models: Synthesis algorithms could become so advanced that some people could become more attractive versions of themselves and create fake model personas that make a lot of money on websites like Twitch. Our economy and education system are probably unprepared for the scale of disruptions we may see.