Will google destroy ChatGPt ?

Will Google destroy ChatGPT ?

I recently read? articles , watched? videos? and experienced some of the AI.

Here are a few reasons I believe google is not going to sit down and look people call their search engine outdated ?

Below I will discuss some of the advance AI from google

chatGPT-3 is based on google’s transformer? architecture that google introduced around six years ago.

Sparrow is similar to ChatGPT and Sparrow is the weakest of the three languages model from google and based on the chinchilla model.

Chinchilla model? is based on a 70 billion parameters language model that outperforms much larger models including gopher.

Because Google's advanced training techniques are outperforming GPT-3 with 170 billion parameters on every possible benchmark test sparrow.

Sparrow? is a fine conversion of chinchilla just like ChatGPT is a fine conversion of GPT 3.5 except sparrow has the advantage of being trained to deliver? evidence for its answer and provide links when users ask for it. Only this difference is a huge win for Sparrow over ChatGPT because not knowing how to validate the chatGPT answers? are? holding lots of users back. I am personally experiencing it.

?Sparrow is worked by DeepMind, which is one of the subsidiaries of alphabet. They have been at the forefront of research for more than a decade now.

Deepmind is the same company that made alphago , alphacode and alphaMind to change the entire AI research landscape each time.

On February 3, 2022 report DeepMind's AI programming tool AlphaCode tested in top 54% of human coders by Bob Yirka ? , which was above the median in competitive programming? against challenges created by code force.

alphaTensor , which is an AI for efficient matrix multiplication, which is another AI that will make other AI more efficient.

These AI listed above are on the same level like ChatGPT or might be a little bit above chatGPT but this is not even Google's best AI model yet.

I am not sure if you’ll remember the google engineer Blake Lemoine ? Lemoine who was fired a few months ago from violating data security policies at google ?

Blake Lemoine is a regular person but he is not a regular AI developer.

He was an AI expert that specialized on working to push? Lamda to its limit.

Blake Lemoine at Google was doing research on how different AI systems can be biased and how to remove bias from those systems. It was testing how bias is with respect to gender , ethnicity and religion.

Blake Lemoine ‘s every day job was to work with the most advanced AI models

On twitter Blake Lemoine tweeted this

“ I love how people are talking about how ChatGPT is gonna replace google while google just quietly uses a more advanced dialogue system behind the scenes.

ChatGPT is a Big step in the right direction but there still few years behind google “

In Addition, Blake Lemoine quoted? “model like ChatGPT is just a module inside lambda “ he also added “ Mina is the language model inside Lambda similar to ChatGPT”

As example, he states? something like this “ if you have a conversation with ChatGPT and ask for example a movie and let’s say that movie is not inside the model , then it is the end of the conversation since it? won’t be able to provide you that data “

Lamda on the other hand could query the entire? web and return your data.

Lamda has a lot of modules . Moreover , Lambda has been trained on dialogue , which makes it a lot more efficient. Lambda’s answers are simple, short and much realistic.

Lambda is not even the most advanced AI model.

Palm is the most advanced AI model from Google trained with 540 billion parameters in the world that was trained on more that 6000 TPU chips and data consists of social media conversation, filter web pages, books,? code and more . This model is multilingual and also helps in code completion and more .It outperformed all the previous AI models such as GPT-3, chinchilla , lamda , etc? on every measure.?

Palm outperformed the average human success rate in solving the most challenging intelligence-based tasks.

I personally think Google AI might be better but Google is afraid of hurting their search business with AI and there’s no proven model yet on how to profit from something like ChatGPT.

I really like all this AI world but I am very afraid about the future since I believe AI could? hurt humanity . I also think it is time to really pay attention to AI before it is too late.

Let me know what you think about AI , ChatGPT , and the upcoming google? release for GPT’s competition.

Happy night monday.

#chatgpt #googleai #openaichatgpt #googledevelopers #linkdin #aiengineer #lamda #sparrow #chatgpt3 #mlengineer