Huge announcements in the world of AI today from Adobe, Elon Musk, the NBA, and Meta.

Here's a rundown on everything you need to know:

1. Adobe adds Firefly to video editing

Edit videos using simple text commands, plus:

-AI color corrections/enhancements
-Animated text and motion graphics generation
-Matching B-roll footage
-Music and sound effects
-Auto-generated storyboards

2. Elon Musk reveals TruthGPT

In a recent interview, Elon Musk said he will start a maximum truth-seeking AI called ‘TruthGPT.’

The AI will try to understand the nature of the universe.

3. ChatGPT got featured in a new NBA commercial with Jack Harlow

ChatGPT has gone mainstream, fast.

Everyone is talking about ChatGPT and Generative AI.

Beyond the hype and novelty applications, its important to start thinking about how it impacts our economy and our businesses.

This is a good way to get your thinking started

This roadmap is a good place to start.

UK’s new technology department to tackle country’s metaverse, Web3 strategy

The U.K.'s newly formed Department of Science, Innovation and Technology will advance the country's metaverse and Web3 strategy.

The country’s 2023 Spring Budget, released in March, said the government wanted to “lead on the future of web technology, sometimes known as Web3 or the metaverse.”

The new tech department, set up in February, will lead the charge in this work.

The department’s work will not focus on specific technologies like blockchain and virtual reality but will instead look at potential economic growth opportunities, investment and business models associated with concepts including the Metaverse and Web3, along with implications for regulation, according to the source.

NVIDIA just released a very impressive text-to-video paper

Video Latent Diffusion Models (Video LDMs) use a diffusion model in a compressed latent space to generate high-resolution videos.

Here's a brief overview of how it works:

1. Pre-train image LDM on a dataset of images.

2. Turn the image LDM into a Video LDM by adding temporal layers to model video frames.

3. Fine-tune the Video LDM on encoded video sequences to create a video generator.

4. Temporally align diffusion model upsamplers to generate high-resolution videos.

5. Validate Video LDM on real driving videos of 512x1024 resolution, achieving state-of-the-art performance.

6. Apply the approach in creative content creation with text-to-video modeling.

Paper: https://arxiv.org/abs/2304.08818

New smart glasses tell u what to say on dates using GPT-4

A team of crafty student researchers at Stanford University have come up with a pair of smart glasses that can display the output of OpenAI's GPT-4 large language model — potentially giving you a leg up during the next job interview, or even coaching you during your next date.

The device, dubbed rizzGPT, offers its wearer "real-time Charisma as a Service" (CaaS) and "listens to your conversation and tells you exactly what to say next," as Stanford student Bryan Hau-Ping Chiang explained in a recent tweet.

The glasses were made using a monocle-like device that can be snapped onto practically any glasses, built and donated by Brilliant Labs. It features a camera, microphone, and a high-resolution display that can output text generated by GPT-4.

OpenAI's speech recognition software, Whisper, allows the glasses to feed speech directly to the chatbot, which can generate answers in a matter of seconds to its wearer.

This paper provides a comprehensive analysis of GPT-4, an AI chatbot, and its potential applications, benefits, limitations, and risks in the field of Medicine.

The authors discuss the potential of GPT-4 in medical diagnostics, patient care, and medical education, as well as the ethical, legal, and security concerns surrounding its use.

Key messages:

1. GPT-4 has shown promising results in medical diagnostics, patient care, and medical education, but several challenges need to be addressed.

2. Ethical, legal, and security concerns are crucial factors that need to be considered when implementing GPT-4 in healthcare.

3. Ensuring data privacy and addressing biases in the AI model are essential to ensure the safe and responsible use of GPT-4 in medicine.

This paper is worth reading because it offers a well-rounded understanding of GPT-4's potential in revolutionizing Healthcare while also highlighting the challenges and concerns that need to be addressed to ensure its safe and responsible use.

EU lawmakers voted 517 in favor of MiCA, making it the first major jurisdiction in the world to introduce comprehensive crypto laws. It also voted 529 in favor of a separate law called which would require crypto operators to identify their customers.

Amazon to power 3M's clinical AI development

3M Health Information Systems announced this week it is partnering with Amazon Web Services to use Amazon’s machine learning and generative AI models to improve 3M products, including its AI medical scribe product, Fluency Align. 

The news comes on the heels of Nuance announcing plans to integrate GPT-4 into its DAX medical scribe product.

The race to integrate large language models like GPT-4 into health care products has swiftly sparked concerns among experts who say it’s not clear yet whether the tools are accurate or reliable enough for use in medicine. 

Apparently, Google Brain team just merged with DeepMind.

Google Brain + DeepMind = Google DeepMind

⚡️Yokosuka becomes Japan's first city to use ChatGPT for administrative tasks

The city of Yokosuka in Kanagawa Prefecture has a few claims to fame: It’s home to a major U.S. naval base, it’s the birthplace of former Prime Minister Junichiro Koizumi, and it lends its name to a local variation of Japanese curry.

On Thursday, it staked out a new claim as the first municipality in the country to use ChatGPT in its municipal offices.

Roughly 4,000 employees at Yokosuka’s municipal government office began using the artificial intelligence-powered chatbot, which was created by OpenAI late last year, for a one-month trial in efforts to improve operations.

Greg Brockman’s TED talk demoed ChatGPT plugins.

Right away he had:
- ChatGPT create a meal idea
- DALL-E generate an image of it
- Retrieval save the chat
- Instacart add the ingredients from memory to the cart
- Zapier create & send a Tweet about it

This is the future.

AI talent is flowing from Silicon Valley to biotech (It used to be the opposite). It's going to be fun to watch what happens in that space.

New capabilities in Bard will help programmers and software developers with code generation, debugging and code explanation. It’s an exciting next step in how generative AI can accelerate innovation across industries.

Amazing! New paper shows biological age rapidly increases during stress & reverses when the stress is relieved.

In parallel to recent developments in like GPT-4, a group of scientists has recently proposed the use of neural tissue itself, carefully grown to recreate the structures of the animal brain, as a computational substrate.

Stablecoin transaction volume looks incredibly strong, even in the current market environment. Whatever people may say, the market data is clear.

People and businesses are adopting stablecoins because they offer them utility and value.

If the next version of GPT is a leap in performance as GPT-4 was fine tuning becomes a thing of the past

Current GPT-4 can go up to 32k context length, what would it mean if the next version did 64k or higher?

Your vertical domain specific model would be one prompt away.

People still want ownership over their models but the reality is that the next GPT could make it hard to justify fine tuning a model of lower quality.

Attention all neuromodulation engineers! Clarity is hiring a talented and experienced Neuromodulation Engineer to join their team.

In this role, you will work on developing and optimizing cutting-edge neuromodulation technologies to improve the lives of patients suffering from neurological disorders.

They are seeking someone with a passion for innovation, a keen eye for detail, and a track record of success in the field. If you're ready to take your career to the next level, apply now and join their dynamic and innovative team.

Global gaming report Q1'23.

Gaming public companies saw a strong recovery in Q1’23, the sector outperformed the broader market indices, and we anticipate M&A activity to strongly rebound in the 2nd half of the year.

1. The Drake Star Gaming Index, an equal weighted index of select top 25 gaming companies with ~1B or more market cap. The index was up 12% in Q1 and LTM P/E and EBITDA multiples (median) expanded by 21% and 12% respectively. 

2. Q1 was relatively quiet with 43 announced deals compared to 76 deals in Q4’22. While the majority of the deal values were undisclosed, several of the top buyers were active including Savvy Games Group / Vindex), NetEase (SkyBox Labs) and Embracer Group (Captured Dimensions and ALL CAPS ApS).  

3. Private financings stayed steady with over 200 deals (212 deals last quarter), raising ~$1.3B. Most active VCs list for Q1 was led by BITKRAFT Ventures, followed by Andreessen Horowitz, Play Ventures and Griffin Gaming Partners. 
 
2023 Outlook  

Q2’23 has started on a great note with Savvy Games Group entering the mobile gaming and publishing space with the acquisition of Los Angeles based Scopely for $4.9B and SEGA announcing the acquisition of Rovio Entertainment Corporation for $776M.  

While continue to see a steady flow of smaller to mid-sized M&A deals and several large deals, anticipate M&A activity to strongly rebound in the 2nd half of the year. Also expect several listed gaming companies to be acquired by PE firms this year.  

Expect a healthy number of early-stage financings and significantly less later stage rounds. Hot investments segments will be tools, AI, VR/AR.

5 latest top AI research papers

- Scaling transformer to 1M tokens
and beyond with RMT.
Introduces a new approach that uses recurrent memory to extend the context length of BERT, while maintaining high accuracy.

- Efficient Methods for Natural Language Processing: A Survey.
Highlights the need for efficient methods in NLP and provides guidance and promising research directions for achieving similar results with fewer resources.

- Alignment in LLMs
The paper proposes a theoretical approach called Behavior Expectation Bounds (BEB) to investigate alignment's inherent characteristics and limitations in LLMs.

- Anxiety in LLMs
Inducing anxiety in large language models increases exploration and bias. Demonstrated that the behavior of GPT-3.5 can be predictably influenced by emotion-inducing prompts.

- Think Before You Act
Proposes a new method for unifying language reasoning with actions in a single policy.

Can BCI-controlled drones help treat ADHD in children?

Researchers from the Universidad de Guadalajara and Tecnológico Nacional de México created a cognitive training program called CogniDron-EEG that uses an EMOTIV EPOC mobile EEG headset.

The system combines feedback from the drone, sound, speech, and a child's therapist with the hope of providing effective, customized training for children, both with and without ADHD.