Future job prosperity, AI for fun, bilingual BCIs, AI-generated TV shows, and more

📖In this issue

• The case for future job prosperity: 13 reasons

• PwC gives ChatGPT Enterprise to 100,000 professionals

• Anyone can create their own AI-generated TV series

• LLM performance on “hard prompts”

• Mapping the evolution of top-tier LLM competition

• Katri Manninen on AI in screenwriting, consciously choosing AI and human roles, creative workflows, and content automation

• Is AI sentient or could it be?

🤸 Future of work

The case for future job prosperity: 13 reasons

Many are crying doom about the future of work. I believe that a highly prosperous future of job is possible. It is even likely, on condition we do the right things today.

In order to create that future, we must believe it is possible. In this mini-report I have summarized 13 strong arguments that a prosperous future of jobs is possible.

Have a look and please let me know how I can strengthen the arguments by pointing out the weaknesses and what’s missing.

👩🤖Humans + AI update

PwC gives ChatGPT Enterprise to 100,000 professionals

“As the technology stack gets better, we can buy versus build more things. We can then focus more on outcomes, transformation, workflow, use cases, and business process, and less on assembling APIs to build an experience for our employees,” said PwC’s Bret Greenstein.

OpenAI signs 100K PwC workers to ChatGPT’s enterprise tier as PwC becomes its first resale partner

Brain computer interface enable bilingual speech

“For the first time, a brain implant has helped a bilingual person who is unable to articulate words to communicate in both of his languages. An artificial-intelligence (AI) system coupled to the brain implant decodes, in real time, what the individual is trying to say in either Spanish or English.” 

First ‘bilingual’ brain-reading device decodes Spanish and English words 

Anyone can create their own AI-generated TV series

“The next Netflix won’t be purely passive. You will be at home, describe the show you’d like to watch and within a minute or two start watching, Finish a show that you enjoy and make new episodes, and even put yourself and your friends in episodes – fighting aliens, in your favorite sitcom and solving crimes.”

Fable’s Showrunner showcases Netflix of AI with user/AI-generated TV shows

GenAI helps artists to protect and scale their work

“We enable artists, people who can draw, and illustrators to train their own AI on their artwork. They then get an algorithm that can generate images in their own style. This algorithm belongs to them and all the inferences belong to them.”

Exactly.ai secures $4M to help artists use AI to scale up their output

LLM performance on “hard prompts”

LMSYS has analyzed their chatbot leaderboard on changes in performance on “hard prompts” (requiring specific knowledge, complex, problem solving etc).While some models have comparable performance, Gemini 1.5 Pro and some Llama-3 models perform significantly worse on hard prompts, while models such as GPT-3.5-Turbo-1106/0125, Claude-2.1, and Phi-3 do considerably better.

Introducing Hard Prompts Category in Chatbot Arena

🔥Hot news in AI

• Google responds to the derision over its AI Overview feature recommending glue on pizzas - 9to5 Google

• Mistral releases Codestral, a lightweight high-performance coding model that covers 80+ programming language - Mistral AI

• OpenAI releases ChatGPT Edu for universities for free or low-cost access - Engadget

• Meta’s open source approach to AI has made Mark Zuckerberg popular again in Silicon Valley - New York Times

• A chart of relative performance of LLMs reveals the evolution of competition in the space - Peter Gostev

🎙️Last week’s podcast episode

What you will learn

• Exploring Katri Manninen’s journey from screenwriting to AI

• Using AI to handle repetitive and formulaic tasks

• Maintaining human creativity and originality with AI

• Automating content creation workflows efficiently

• Enhancing cognitive processes and ideation through AI

• Ethical considerations in the use of AI for creative work

• Future possibilities of AI in amplifying creative potential

💡Resources and insights

Is AI sentient or could it be?

In Time magazine Fei-Fei Li, head of Stanford’s Human-Centered AI Center, argues that AI cannot be sentient because it does not have subjective experience.

While I don’t think AI is sentient and won’t be for a long time (if ever), I disagree with this argument. I believe the debates and disagreements over whether AI is sentient will intensify without end, as there is no way of proving it either way.