5 Alternatives to ChatGPT

Hallucinations. Bias. Security and privacy concerns. Lawsuits.

As the uses of ChatGPT in the workplace expand, so do the risks it poses to firms. Some companies have already had proprietary data and other information inadvertently made public after uploading it to the platform. Others have been sued over data privacy, copyright infringement, or bias. Many more have been embarrassed for failing to double-check incorrect information or called out because of ethical concerns around job loss and social impact.

“Most boards and leaders don’t understand the risks of generative AI,” says Chris Cantarella, global sector leader in the software practice at Korn Ferry.

To be sure, many corporate leaders think of ChatGPT and generative AI as synonymous. But since its release last November, ChatGPT has spawned literally thousands of other generative AI apps and platforms—more 17,000 startups launched this year are focused on AI, an increase of 38% over 2022. “There are so many more alternatives to choose from now,” says Cantarella.  

There are, of course, the large language models that compete directly with ChatGPT, such as those from Google, Microsoft, Meta, and others. But experts say the next wave of generative AI is moving away from generalist platforms to targeted applications for specific functions or purposes. Firms can accelerate deployment, more effectively train employees, and realize productivity gains faster with more specialized AI tools tuned for different uses. Perhaps more importantly, they can improve accuracy and limit risks because the AI will learn from more precise data, ultimately providing higher quality, more specific answers, says Michelle Seidel, a senior client partner in the global technology practice at Korn Ferry.

Here are five alternatives to ChatGPT our experts say are attracting attention among business users.

For coding

One of the best examples of specialized generative AI is GitHub’s CoPilot. Unlike ChatGPT, which can generate code to guide developers, CoPilot helps them write original code and integrate it into operations. The AI tool is also designed to learn the individual software developer’s coding style over time. The distinction is subtle but significant for software developers, who say the code CoPilot produces runs better and with less errors.

For creating images and graphics

Still in beta testing, Adobe’s Firefly generative AI is winning rave reviews from graphic designers, illustrators, photographers, and other content creators who have been granted access. The AI generates images based on text-based prompts, allowing users to create new designs, highlight, or change image aspects without doing any editing. The feature creators love, says Paul Fogel, sector leader, professional services, in the software practice at Korn Ferry, is the fact that all work used by the AI engine is credited and the creator is compensated.  

For taking meeting notes

It’s the part of the meeting every employee dreads—the job of typing up the meeting notes. But with Otter AI Chat, meeting notes are taken and summarized in real-time based on what participants say. The AI isn’t just a passive recording device, however. It acts as a meeting participant, answering questions, collaborating with attendees, and generating meeting-specific content on-demand.

For crafting better-sounding text

Using AI to mimic a brand’s unique tone and voice, Jasper.ai was built to help advertisers, marketers, and communications professionals create blog posts, social media content, press releases, website copy, and more. The tool allows users to share and collaborate on the same document, which ChatGPT does not.

For a direct ChatGPT alternative

Developed by two siblings who previously worked at OpenAI, Claude is a generalist platform that competes directly with ChatGPT. Cantarella says Claude’s processing power and knowledge base are more advanced than ChatGPT’s, for instance. The developers built that platform using a practice called constitutional AI to produce “less harmful” results. The idea is to train the AI to generate responses that are socially and ethically responsible by adhering to a prewritten framework for the answer. So far, however, the results have been mixed, with reviewers who have been granted access to the beta version of Claude saying it still generated biased, false, or harmful content.

Learn more about Korn Ferry’s Business Transformation capabilities. 

Read more This Week in Leadership articles