If you have an AI idea that you want to get funded, now is your time to try.
VCs have been pouring billions into AI startups for years, and since ChatGPT went mainstream in December, the acceleration has been reaching a fever pitch.
ChatGPT was the fastest consumer product to hit 1 million users, and interest in it has increased - most recently in China and Southeast Asia.
There's a meme now that all you have to do to raise money from VCs is build a UI wrapper around a ChatGPT function - like an app to give you customized recipes or help you code. Even Snapchat is releasing a new chatbot with ChatGPT on the backend.
But as Balajis put it.
Satya vs. Sergey
Companies are feeling the competition. So much so that Google co-founder, Sergey Brin, came out of the shadows of his private islands and submitted his first changelist (request for code access) in years to none other than LaMDa - Google's natural language chatbot.
It makes sense, given what Microsoft's Ceo, Satya Nadella, said.
I want people to know that we made them dance.
That brings us to Bing.
For those of you unfamiliar with Bing's background, Trung Phan laid out its history in his newsletter.
The future of Bing (and maybe Clippy) is uncertain, but its present has recently become a topic of interest. Bing now incorporates variations of the technology used in ChatGPT to answer users' questions and provide a new, arguably revolutionized, browsing experience. Of course, there are variations of this already in the wild - the best alternative I know of is You.com, which has been incorporating a chat search feature for months. At times, it's returned better results than ChatGPT in my use cases.
As you can see with Notion's AI, GitHub's co-pilot, Replit's AI, Meta AI, and SnapChat's new chatbot - to name a few, there will be a lot of competition in this industry.
So why is OpenAI valued at $29 bil?
Let's not get into that.
Let's focus on what could happen next.
Public Perception
First, we'll need to see how the public is reacting to the advances in AI - here's Astrid's take.
The idea of fearing an imminent AGI threat and going out and spending all your money seems extreme, but it's real.
AI advancement will have many more unintended consequences, but what if it slows down? Standard expertise acquisition usually follows an S-curve where there is a period of stagnation after sudden advances.
ChatGPT has passed many difficult exams that are meant to represent subject matter expertise. ChatGPT is a subject matter expert at times, but it's wrong far too often to be taken into serious consideration as a viable approach for search or medicine.
A known hallucination rate hovers between 5-10%, and some other bells and whistles have raised eyebrows.
Ted Goia breaks it down in his newsletter.
Given the nature of S-curves, some speculate that the progress AI has made recently will be the most it experiences for a while (until the next leap forward in the sweet spot region of the above graph - maybe quantum computing gets us there?). If this stagnation occurs, it will be a repeat of what happened during the AI winters of 1974-1980 and 1987-1993 - booms and busts in both technological advancement and investment.
Right now, we're in a boom. Will it end differently this time?
Bing requires Edge
To access the new Bing, you currently have to use the Microsoft Edge browser, which is dull compared to shiny new toys like Arc browser. I've been exclusively using Arc for over a month, and it's my new default - but I can't use Bing's chat feature.
There's also a mobile app for Bing, and it's functional but stressful. I've been an Apple person since early college, and I don't like using Windows outside of an emulator - so being forced to use a Microsoft product in an Apple environment is not ideal. Also, nearly no one uses Edge or Bing - but that is interesting in the grand scheme of Satya's strategy.
So far - anecdotally, Bing's chat feature has been underperforming both ChatGPT and You.com. Poe, brought to you by the team behind Quora, has also returned more valuable results. Because of the wonkiness that Ted detailed in his newsletter, Microsoft placed a 5-response limit for each conversation.
Wonkiness
Bing has returned strange things to me like "I would rather us not have this conversation anymore 😔" instead of giving me a wrong answer - as ChatGPT, You.com, or Poe would.
Note the word return - I am not going to anthropomorphize these chatbots. It's not responding to me or talking to me. Prompts are the same as binary code to these things, and I view their entire existence as a simple function with emergent properties.
On another note, I am creeped out by Bing's use of Emojis - look at the screenshot below.
It was taken from this thread made by Juan Cambeiro.
And then there's this one 😳.
The Underdog
In the above interview, Satya made it clear that improving Google's search model has been daunting. They are monolithic and, until recently, have had no competition. He sees ChatGPT and Bing as the first real competitor to Google's prowess. Combine this with the fact that Nvidia's CEO went on the record to say that their chips will be 1 million x the power of current AI models in the next decade. ChatGPT runs on an estimated 10,000 Nvidia GPUs.
Things could get dicey for Google if this is true. Satya directly called Google out and will have billions spent competing with them. With Edge and Bing only amounting to a fraction of Google's market share, what do he and Microsoft have to lose?
Every good underdog story has a Goliath; this time, its Google.
Although, one could argue all of the recent progress in AI wouldn't have happened without Google. Transformers are considered game changers in the application of AI models, and all of these chatbots were trained on networks that incorporate transformers… and Google's research department is where the transformer was first developed.
Google may speed up its historically slow software development cycle and blow everybody out of the water. Or maybe Apple will, or Tesla, or Anthropic, or Buzzfeed?
Inspiration
In 1962, Kennedy talked about landing on the moon within a decade.
Unfortunately, he didn't get to see it, but the US landed on the moon in 1969. Then, all of a sudden, Sci-Fi became an exciting topic in mainstream culture, and space blockbuster after space blockbuster was produced in Hollywood. You start hearing about Star Wars this and Star Trek that and Alien this and Stanly Kubrik's 2001 Space Odyssey that, etc. UFOs are suddenly spotted all over the US, and STEM subjects become the academic system's focal point.
The generation before the moon landing probably thought about space less than the generation after it. The same will be the case for radical improvements in AI like ChatGPT. At first, the most significant societal changes will be inspiration or fear - similar to Sergey Brin's reaction. Then unintended consequences that no one can predict will start happening for better or worse.
I hope people don't start marrying chatbots.
We’ll end with a photo of OpenAI’s co-founder, Sam Altman, posing alongside Grimes and Eliezer Yudkowsky.