As the AI arms race between tech giants heats up, OpenAI is taking a new approach to protecting its technology and brand. The company has filed for a trademark on ‘GPT’ with the United States Patent and Trademark Office (USPTO). While simultaneously, it is implementing brand guidelines to maintain control over its image and technology usage.
AI Giants Battle for Supremacy
Google has been striving to outpace the Microsoft and OpenAI partnership in the AI realm. Google is planning to do so by merging DeepMind and Google Brain into a single entity. OpenAI, already ahead in the race, is attempting to solidify its position by trademarking ‘GPT.’
Also Read: Google VS Microsoft: The Battle of AI Innovation
OpenAI’s Trademark Application
OpenAI submitted the trademark application for ‘GPT’ in December 2022, but the approval process may take 4-5 months. The company recently filed a petition urging the USPTO to expedite the process. This is because numerous apps bearing the GPT name have emerged. Jefferson Scher, a partner in the intellectual property group of Carr & Ferrell, informed TechCrunch that the application is still pending.
Establishing Brand Guidelines
OpenAI has published brand guidelines on its website in response to the delay. Open AI wants to ensure developers building AI or GPT products don’t falsely claim to masquerade their creations as OpenAI. The company aims to preserve its popularity and credibility with this move.
By filing for a trademark rather than a copyright, OpenAI seeks to prevent others from releasing technology under their name while still allowing the use of their technology with proper attribution. The brand guidelines outline the appropriate terminology when building products powered by OpenAI’s technology. For instance, a “MeowlyticsGPT” product should be renamed “Meowlytics powered by GPT-4.” Plugins must specify their compatibility with ChatGPT instead of claiming affiliation with OpenAI.
Learn More: Everything You Need to Know About ChatGPT
Controversy and Potential Backlash
However, Reddit and HackerNews users have criticized OpenAI’s trademark application, arguing that the company didn’t introduce GPT nor coin the term. Furthermore, they contend that the trademark application comes too late, as GPT-1 was first introduced in 2018, and the phrase “generative pre-trained” can be traced back to earlier Google research papers.
Regardless of the company’s original intentions, the trademark application could serve to prevent cloning or signal larger ambitions under Sam Altman’s leadership. OpenAI has registered AI.com to redirect to ChatGPT, making a strong statement in the AI market.
Competitors’ Opportunity to Respond
If the USPTO accepts or even considers OpenAI’s trademark application for ‘GPT,’ it will enter an ‘opposition period’ during which competitors like Google or Meta can raise concerns. This may allow Google to catch up in the AI competition.
Also Read: More Profound than Fire or Electricity: Google CEO Sundar Pichai on AI Developments
Our Say
The trademark application and resulting name changes for GPT-based products could negatively impact developers who have built their products using OpenAI’s APIs. The company must rebrand these products if the authorities approve OpenAI’s trademark application. This could make them less appealing to customers.
This move might also reduce the number of people using OpenAI’s APIs. The company aims to maintain control over its technology while allowing for some usage. The trademark application’s outcome remains uncertain. But OpenAI is determined to protect its brand and technology in the ever-evolving AI landscape.