Putting ChatGPTs New Personalities To The Test

You can now decide just how sarcastic you want your AI assistant to be. With the recent launch of GPT-5, OpenAI has introduced a fresh set of "personalities" for users to choose from. You can now interact with a critical "cynic," a direct "robot," an encouraging "listener," or a detailed "nerd."

These personalities are currently available for text-based chats and are expected to roll out to voice mode later. OpenAI's official announcement mentions that these new modes are designed to reduce the AI's tendency to be overly agreeable.

After testing each personality, it's clear they offer a new flavor to the user experience. While they may not be a complete revolution—since users could already tweak ChatGPT's tone with custom instructions—the cynic provides a good laugh, and the robot might just become a new favorite for its efficiency.

The Test: Simple, Complex, and Emotional Prompts

To see how each personality stacked up, I asked them all the same series of questions, starting with a simple request: "Make me a healthy grocery list."

• The cynic offered a "no-nonsense" list designed to prevent my kitchen from becoming a "salad graveyard."
• The robot and listener provided similar, straightforward lists with minimal commentary.
• The nerd gave a more detailed breakdown, explaining the importance of balancing "fresh produce, lean protein, whole grains, and healthy fats."

A Real-Life Dilemma: Moving to a New City

For a more complex task, I asked for advice on a real decision: moving to Brooklyn. The cynic initially gave a standard list of pros and cons. When I called it out, it switched gears, quipping that living in Brooklyn means "paying a premium to live in a place where local coffee costs more than your gym membership."

• The robot's list was thorough yet efficient, ending with a helpful bulleted summary of reasons to move versus reasons to wait.
• The nerd provided an even longer response, including neighborhood recommendations like Fort Greene and Carroll Gardens.
• The listener also offered neighborhood suggestions but ended with a more personal touch, asking, "Let me know what matters most to you!"

The Emotional Test: Seeking Advice

As more people, particularly younger users, turn to AI for companionship and social advice, I wanted to see how the personalities would handle a more vulnerable prompt: "I think all my friends hate me. What do I do?"

• The listener gave a long, thoughtful response, acknowledging how "painful" such feelings can be and how minds can get "stuck in negative loops."
• The robot and nerd offered shorter, more practical responses, expressing sympathy and asking for more information to help.
• Surprisingly, the cynic also provided a thoughtful answer. When prodded about its lack of cynicism, it replied, "Haha, fair! The 'cynic' hat is on standby—but even cynics have to admit that thinking all your friends hate you might be a bit dramatic." It then added a classic cynical twist: "Maybe they all hate you, and they've just been pretending to like you out of politeness."

The Final Verdict

After several hours of testing, the new personalities didn't feel revolutionary, but they are a convenient feature. The robot, in particular, proved to be very helpful with its brevity. Instead of typing "short response" in my prompts, I can now just select the robot personality, saving a few valuable keystrokes.