OpenAI is still a new company, but if you haven’t heard of it since its founding in 2015, you’ll know it as a household name soon enough. The company is on the cutting edge of the artificial intelligence (AI) industry. The growing popularity of AI can be chalked up to the company’s increasingly popular ChatGPT chatbot, and its collaboration with many a tech giant to integrate its products across the space. One company deeply familiar with OpenAI is none other than Microsoft (NASDAQ:MSFT), which is using the newest of the startup’s services.
Microsoft is known as a trailblazer in all sorts of tech spaces. The software company is no stranger to new products and services, and it is jumping quickly on the AI bandwagon. Already, it has integrated AI into its Bing chatbot. The results have been mixed, to say the very least. At worst, the bot has been accused of paranoia, rudeness and name-calling. This led to the bot being immediately yanked and rolled back into the proverbial workshop for tweaking.
A soft re-release has been going over fine. But, what’s most interesting about Microsoft’s AI journey is its hefty backing of OpenAI. In 2019, the company invested $1 billion in OpenAI. Part of this deal included powering the company’s AI model. Not having the tech, Microsoft has been building a supercomputer for the startup. It’s also invested over $10 billion in the company since then.
OpenAI, meanwhile, is pushing along improvements to existing products, including ChatGPT. In fact, the company is already rolling out an upgraded version of the chatbot that it says will drastically improve upon the existing AI experience. And as it turns out, users will get access to it before OpenAI’s release.
ChatGPT Upgrade Powers Bing AI, But OpenAI Won’t Reveal Secret Sauce
This week, OpenAI is announcing its ChatGPT upgrade, GPT-4. Developers are touting the model as a vast improvement to an already useful tool. The company asserts that ChatGPT-4 can answer questions faster and more accurately. It adds that it can accept and analyze images, and that its deeper reasoning makes for more robust responses. Most assuring to AI skeptics are claims that the technology is “safer” as well.
As it turns out, Bing users will have a chance to play with GPT-4 before OpenAI releases the product. Microsoft reveals that it has been integrating GPT-4 in its Bing chatbot. It will be releasing the new version as early as this week. Since the announcement, MSFT stock has risen by about 6%.
There are lots of things to be excited about with this upgrade, as the applications for AI only seem to be growing. However, there’s a shade of secrecy with the rollout that is concerning some. Indeed, OpenAI is keeping under tight wraps what exactly makes this model different from the 3.5 one. For the first time since the company’s first GPT release, it is revealing nothing about the data the model utilizes, the techniques used to train the model and more. As experts are pointing out, this likely comes as the company shifts in ideology from a research-based one to a competitive, consumer-facing one.
The GPT-4 model might be exciting and elusive, but it’s also not without criticism. Some point out that the model carries over many of the logical flaws and biased thinking observed in previous iterations of the product. Many attest that the technology is still “untrustworthy” and cannot handle important duties, even with the impressive overhaul.
On the date of publication, Brenden Rearick did not hold (either directly or indirectly) any positions in the securities mentioned in this article. The opinions expressed in this article are those of the writer, subject to the InvestorPlace.com Publishing Guidelines.
