OpenAI Hype Cycle

OpenAI Hype Cycle

Last year, Gartner, in its AI Hype Cycle report, predicted that foundational models (GPT-3.5, LaMDA, PaLM, BERT, etc.) would take about five to ten years before it touches the final stage of mainstream adoption. But, one company shattered them all – OpenAI.

After the launch of ChatGPT (built on the foundational model GPT-3.5) at November-end last year, it took the company only three months to break all records of a product lifecycle. It is pushing tech giants, including GoogleBaiduAmazonMeta and others, to catch up as quickly as possible, alongside the creation of ChatGPT clones and alternatives

Check out the infographics below:

No alt text provided for this image

In such a short span, ChatGPT has already been through the cycle, including emergence, adoption and dispersion, and is now reaching a mature phase of measured expectations, leading to mainstream industry adoption and the next wave of product development and experimentation. A contributing factor to this might be the bot’s meteoric growth, which scaled to 100 million users within 60 days as of January 29, 2023. 

Gartner Hype Cycle explains this best. The hype cycle is a concept created by research firm Gartner to showcase the relationship between time, the maturity of technologies and the expectations associated with it. The cycle generally progresses through five phases before reaching mainstream adoption. 

Delve deeper into these phases here.


Top Stories of the Week >>

Should Sundar Pichai Step Down?

After the Bard launch fiasco, Google has been in the spotlight for all the wrong reasons. This includes various allegations by its employees of accusing Sundar Pichai of rushing the release, alongside its core team working on the critical projects now transitioning to arch-rival OpenAI. Many experts believe that Sundar Pichai's days are numbered. Read more here

This Rust-based Python Linting Tool is Quick as Lightning

A new tool called Ruff has been making noise in the developers' community. So, what is the secret sauce behind its popularity? It is said that Ruff has become one of the primary linter for some of the most vital Python-based projects like Pandas, FastAPI, Apache Airflow, etc. 

Ruff is over 150x faster than Flake8, a popular Python lint wrapper; 75x faster than pycodestyle, the official linter to check Python for the PEP8 Python style; 50x faster than pylint. Read the complete story here.

Machines Are Dreaming, Instead of Learning 

As per Gartner, nearly 60 percent of all data in AI will be synthetic by 2024. While synthetic data would be a vital tool for AI modelling in the future and is definitely on the rise, given the data scarcity problem, there is a flip side to this. Read to find out. 


People & Tech >>

Former Google AI Ethicist Calls Bard and alike ‘Bulls***’ Generators

In an exclusive interview with AIM, former Google AI ethicist Alex Hanna said that many people are freaking out about large language models (LLMs) writing codes, college essays, etc. “The harm is, these things are just bullshit generators,” said Hanna. Full interview available here. 

Quantum Computing Meets ChatGPT 

Can ChatGPT scale to a level of human-level intelligence? Answering this question,  L Venkata Subramaniam, a quantum distinguished ambassador at IBM, told AIM that quantum computing can be used in working with fundamental models like ChatGPT. He said that quantum computing can achieve comparable results to classical AI using less training data and potentially accelerate the training process for AI models. Read more here.

The Timeless Legacy of Tally Solutions 

Established in 1986, Tally Solutions has been one of the oldest known legacy tech players in the country. The company have been actively leveraging emerging technologies, including cloud, data analytics and AI/ML, to bring the best version of its products to the customers. In a recent interview with the director of technology at Tally Solutions, Anjina Murthy tells us how the company's focus on technology and innovation has kept it ahead of the competition and more. 

Read the complete interview here

Fighting the Software Piracy Demon with Neha Singh, Fractal

Piracy is a massive problem globally. Software, media and entertainment (including gaming) are the worst affected due to piracy. Study shows that 37 percent of all PC software is pirated. In other words, one in every three software installed is pirated. So, what is the solution here? AIM caught up with Neha Singh, a client partner at Fractal Analytics, to understand what it takes to measure, regulate and stop software piracy. Interview excerpts here.


AIM Videos >>

Women in Tech 

Who says the semiconductor industry is a man's world? Meet Kripa Venkatachalam, the VP of field engineering at Cadence Design Systems, where she shares her journey in the semiconductor industry and why she wants to encourage more women to take up a career in the male-dominated tech industry.

Join more women leaders like her at India’s biggest women in tech conference, The Rising 2023, held from March 16, 2023, to March 17, 2023, in Bengaluru. 

Register Now!


AIM Shots >>

Microsoft’s all-new Bing turns rouge. Many examples have been circulating on the internet where the chatbot has been acting all lovey-dovey to calling people ‘unreasonable and stubborn’ and more. To cull this, Microsoft has now put a cap on the usage of the platform – limiting it to five questions per session


Recently, OpenAI released a blog where the team explained how ChatGPT works and their decision to allow more users for customisation and more input from the public in the decision-making process.


Indian-American Neal Mohan is stepping up as the new CEO of YouTube, from the chief product officer role. He would replace Susan Wojcicki, who has led YouTube for almost nine years. 


Now you can edit on ChatGPT. A Reddit user who goes by the name shuafeiwang, has developed a Chrome extension called editGPT, that turns ChatGPT into ‘Grammarly’.

Gladio recently released ‘Audio Transcription API,’ built on OpenAI’s Whisper-Large-v2, a speech-to-text API that helps in transcribing 1-hour file in 10 seconds with a word error rate as low as 1 percent. 


Allen & Overy recently introduced a new GPT-3.5 powered, AI chatbot called Harvey, developed by a startup called Harvey, and incubated by OpenAI Startup Fund

JupyterLab Desktop has released a new upgrade called JupyterLab Desktop v3.6. This new update boosts the user experience and makes setting up notebooks on PCs much simpler and faster.

To view or add a comment, sign in

Insights from the community

Others also viewed

Explore topics