The AI competition has come to an end, 8 winners have won, and Seattle may become the outbreak site of the next generation of AI technology

Source: "New Zhiyuan" (ID: AI_era), author: Xinzhiyuan, editor: LRS

In the past year or so, generative AI technology has become popular all over the world, and many new tools have emerged in various fields such as images, texts, videos, and music. If you don’t talk about ChatGPT at work, you will feel like you are out.

At the same time, all top technology companies have also started a battle for talent. After all, large-scale language models, large-scale cluster infrastructure, and AI applications are still high-end technologies that are in the hands of a few high-end talents.

In addition to technology, experience is also important. Developing AI models requires deep knowledge and creativity to create great products.

Recently, a foreign media reporter dug deep into all artificial intelligence-related companies in the world, sorted them according to product quality, number of papers and other indicators, found eight current companies leading the development of generative artificial intelligence technology, and counted the public education of employees Background, previous companies, residence and other information.

In addition to the eight leading companies, the article also counts 11 very potential companies that apply models to actual scenarios and directly connect with users. They may be the leaders of the next AI technology innovation.

In the process of statistics, it can be found that the top talents of large companies are basically poached by more flexible and innovative start-up companies, which makes large companies at a disadvantage in the battle for talents in the wave of AI technology.

Generative AI Eight Leaders

Based on the public employee information, such as school education, advanced degrees, companies they have worked for, working years, patents, publications, etc., eight companies with AI R&D as the core were selected from more than 3,500 companies engaged in generative artificial intelligence business. company.

1. Tome, AI that helps generate speech and visual storytelling

2. Character.AI, you can chat with various generated AI characters

3. Anthropic, a company focused on artificial intelligence security and research, mainly builds chatbots for the working environment

4. OpenAI, the development company of ChatGPT, demonstrated the ability of large language models; also released the DALL-E series models for image generation

5. Hugging Face, building models to perform specific tasks in the enterprise

6. Jasper, a model for creating creative content at the marketing and brand level

7. Stability AI, open source Stable Diffusion, is also one of the earliest image generation models released to the public

8. Midjourney, a self-funded research lab, releases an image generation tool that runs entirely on Discord

In terms of recruiting talents, these eight start-up companies have won the overall victory in the talent competition, hired the highest level engineers and artificial intelligence experts in the same industry, found the positioning of their own products in the market, and obtained enough High financing and valuation.

Companies that package existing models as applications were not considered in the ranking process.

From the perspective of the number of employees, the employees of these start-up companies are the elite among the elite. Most of the companies have less than 200 employees, and the smallest has only 22 employees. It can be said that with the smallest number of employees, they have created the highest productivity in the world.

Through the statistics of the current employees, it can be found that many of them have worked for large companies such as Google and Meta, and even some employees who have switched from OpenAI.

On the other hand, Google, which has always been guided by AI First, has invested a lot of money in the artificial intelligence track, but until now, Google has not come up with an open and available consumer-oriented AI product or service.

Most of the employed employees have educational backgrounds from world-renowned universities, among which Stanford University, University of California, Berkeley, and Massachusetts Institute of Technology are among the top three; and their educational backgrounds are also very international. The School of Artificial Intelligence at the University of Waterloo in Canada ranks fourth, Belgium's KU Leuven School of Artificial Intelligence took eighth place.

Geographically, the Hayes Valley is becoming a frequent location for AI events, meetups, and maker homes.

And with San Francisco also having the worst Covid recovery of any major U.S. urban area, it can be seen that a large number of talents are pouring into the Bay Area, or at least finding new jobs in artificial intelligence companies.

However, the public information does not show which employees in a certain geographic area are remote workers. The statistics in the figure below are definitely lower than the actual values, but they can also reflect the trend.

Next Generation AI Companies: From Research to Application

Although the names of the eight start-up companies appear frequently in major media, there are also 11 companies that are quietly expanding, attracting a large number of high-quality talents to join the team.

1. Pictor Labs, digital pathology (AI-driven histopathology), accelerating clinical research on diseases

2. Endpoint Health, AI-assisted precise treatment of immune-related diseases

3. Otter.ai, audio transcription driven by artificial intelligence, similar to chatbot conference assistant

4. Bria, image and video content with commercial use control

5. WellSaid Labs, text-to-speech dubbing

6. Subtle Medical, Radiology Image Processing Driven by Artificial Intelligence

7. Copy.ai, marketing copywriting via chatbot

8. Lexion, an AI-powered contract assistant for sales, procurement, HR, legal, and more

9. You, a search engine with a chatbot interface

10. OctoML, large-scale language model fine-tuning and operation based on cloud computing

11. Runway, video, image and other creative tools

These companies all have one thing in common: While AI research is at the core, they focus on solving domain-specific problems, or focus on tailoring specific use cases for real customers.

Many of the technologies developed by the eight startups are leading the future direction of artificial intelligence research, and these 11 companies may be the companies that can provide users with real value, tool development, interaction and other services, and may cover everything from healthcare to entertainment. All walks of life.

From the perspective of educational background, the University of Washington jumped to the top and is also home to the Paul G. Allen Center for Computer Science and Engineering, followed by Stanford University and the University of California, Berkeley.

When surveyed for employment background, the University of Washington reappeared, along with Amazon, Microsoft and the Allen Institute for Artificial Intelligence (the research institute founded by the late Microsoft co-founder Paul Allen).

Among the 11 AI companies, Wellsaid, You, and Runway are all headquartered in Seattle, where several top companies and institutions, such as Microsoft, Amazon, Boeing, etc., are also coveting AI talents.

In 2021, a report by the Brookings Institution listed the metropolitan areas of early adopters of artificial intelligence, named "eight large technology centers", including New York, Boston, Seattle, Los Angeles, Washington, D.C., San Diego, Texas. Austin, Texas and Raleigh, North Carolina.

Where is the future?

The pace of technological change is still accelerating. From the invention of automobiles, to airplanes, to semiconductors, human beings have seen the global influence of innovation.

Over time, generative AI technologies will reorganize the way almost everything in the world works and develops, whether it is agriculture, education, or corporate marketing, finance, etc.

In every great change, talents are always at the core.

While fears are always overdone that AI will replace all jobs with robots, generative AI startups need more than ever the specialized skill sets needed to build new features and products, let alone manage all associated Technology, such as running a data center with thousands of GPUs, compressing thousands of bytes of training data.

Work around language models is just beginning, as companies race to train custom models based on the technology and refine those already in open source use.

For this reason, the number of stars related to artificial intelligence projects on Github can also show a certain technological trend. Anyone can contribute and learn how to develop artificial intelligence applications. Recently, it has begun to explode, especially OpenAI in After ChatGPT is released to the public on November 30, 2022.

In the first quarter of 2023, there will be approximately $1.7 billion in deal value generated in the artificial intelligence space, which does not include the $10 billion investment in Open AI announced by Microsoft in January.

While it's still unclear how the computer industry of the future will develop, the AI revolution may be the fastest-growing technological boom and transformation in human history.

Reference materials: