One Year Ago, It Was Only ChatGPT — Now Large Language Models Dominate the Scene

One Year Ago, It Was Only ChatGPT — Now Large Language Models Dominate the Scene

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More.

When OpenAI launched ChatGPT a year ago, the concept of an AI-driven personal assistant was new to many people. We had seen chatbots before, but they were limited to pre-programmed responses. Few believed in the possibility of a conversational companion capable of discussing various topics and assisting with tasks like creating travel plans, offering insights on complex subjects, or writing detailed articles.

Fast forward to today, ChatGPT, despite some company drama—including Italy’s brief ban and Sam Altman’s temporary ouster and return—remains a popular choice among AI assistants for millions of online users. Available on both web and mobile platforms, ChatGPT helps with numerous tasks and sees billions of interactions.

In October 2023 alone, ChatGPT had nearly 1.7 billion visits across mobile and web, with 193 million unique visitors and an average visit duration of about 8 minutes, according to SimilarWeb. To put this into perspective, it’s significantly more than the engagement seen by services like Zoom (214 million visits) and Google Meet (59 million visits).

Beyond introducing conversational AI to millions in just a few months, ChatGPT has also sparked growth in the broader AI ecosystem. Generative AI quickly became a hot topic, leading to the launch of numerous consumer-oriented tools for creating text, music, videos, and code. For both chat and code, several services like Github Copilot and Perplexity AI have utilized fine-tuned versions of GPT models, which power ChatGPT. Many businesses have also used OpenAI’s APIs for their internal applications and custom GPTs for specific tasks like data analysis.

Importantly, this AI boom has also driven many startups and tech giants to develop their own large language models (LLMs) and general-purpose chatbots. While these may not be as popular as ChatGPT yet, they significantly democratize the space and ensure that OpenAI isn’t the only player.

Following OpenAI, companies like Anthropic and Google launched their own conversational assistants, Claude and Bard. Anthropic, founded by ex-OpenAI employees, emphasizes safety and ethics, training Claude as a “constitutional AI” which follows specific rules and improves with automated feedback. Bard, built on Google’s Pathways Language Model 2, integrates with Google Search to provide detailed, context-rich answers. These two are seen as ChatGPT’s biggest competitors.

But the competition doesn’t stop there. Other chatbots, such as Inflection AI’s Pi, designed to be more personal and conversational, and Cohere’s enterprise-focused Coral, have also emerged. Meta’s open-source Llama 2 model has spurred further development and fine-tuning of various AI applications. Recently, Reka launched Yasa-1, which understands and processes text, images, audio, and short videos. Elon Musk’s xAI introduced Grok, which adds humor and sarcasm to its real-time information retrieval.

Even though all these assistants aim to help with tasks like general search and text summarization, they are constantly evolving. Their current capabilities and limitations might change significantly in the near future. For instance, ChatGPT initially launched as a text-based assistant but now accepts voice commands and images and even responds with its own voice like Alexa.

In addition to the major AI assistants, several market-specific versions have been developed to cater to different industries, audiences, languages, and unique needs. Notable examples include Naver’s HyperClovaX in South Korea, China’s Ernie and DeepSeek, and AI assistants like Poro and Nucleus for the agricultural sector. India’s Reliance Industries is also developing a generative AI assistant for the country’s diverse dialects, supported by Nvidia’s computing infrastructure.

With numerous companies working on delivering effective conversational AI and even more on the horizon, the AI race is far from over. As the underlying models improve and chatbots become better at providing natural and relevant responses, the competition is expected to tighten, pushing the boundaries of AI further.

On its first birthday, ChatGPT remains a leading player in the AI sphere. However, it will be fascinating to see how it maintains its leadership in the coming years and contributes to the development of safe and responsible artificial general intelligence (AGI). According to Precedence Research, the global conversational AI market is projected to grow nearly 24% annually, surpassing $86 billion by 2032. Will LLMs become commonplace, with each industry or even every company having their own version? Time will tell. Check back in a year to see how things have evolved.

Stay in the know! Get the latest news in your inbox daily.