When I logged into OpenAI’s website last year to continue testing the new version of ChatGPT powered by GPT-4, the chatbot didn’t try to sabotage my relationship, write my emails, or unleash my creativity—it simply didn’t work. Demand was high that day, and the company was experiencing occasional outages. Greg Brockman, an OpenAI cofounder and president, was upfront about the model’s imperfection in a livestream around the same time. He also reminded listeners that they were not without blemishes themselves.
Generative AI remains a focal point for many Silicon Valley developers after OpenAI’s transformational release of ChatGPT in 2022. The chatbot uses extensive data scraped from the internet and elsewhere to produce predictive responses to human prompts. It was previously powered by the GPT-3.5 language model. While that version remains online, an algorithm called GPT-4 is also available with a $20 monthly subscription to ChatGPT Plus.
If you’re considering that subscription, here’s what you should know before signing up, with examples of how outputs from the two chatbots differ.
The core service you pay for with ChatGPT Plus is access to GPT-4. You’re not buying an unlimited number of prompts, though. Subscribers are currently limited to inputting 40 prompts to GPT-4 every three hours. After hitting the prompt limit, users can always switch to the GPT-3.5 version. (In addition, you can experiment with a free version of GPT-4 by signing up for Microsoft’s Copilot chatbot.)
A ChatGPT Plus subscription also unlocks OpenAI’s newest software features. For example, Dall-E 3 is available inside the bot to generate images for you via text prompts. Subscribers can also use ChatGPT’s Bing integration to browse the web in real time for information that may be relevant to their prompts.
Another core aspect of ChatGPT Plus is the GPT Store. Here’s where you can access versions of OpenAI’s bot that have been customized by the community with additional data and parameters for more specific uses, like coding or writing help. You can even try out a unique bot that’s based on my writing for WIRED.
Many aspects of GPT-4 remain opaque. OpenAI has not shared many details about GPT-4 with the public, like the model’s size or specifics about its training data. With that in mind, are answers from the GPT-3.5 and GPT-4 models noticeably different? OpenAI’s website reads, “GPT-4 excels at tasks that require advanced reasoning, complex instruction understanding, and more creativity.”
It seems like the new model performs well in standardized situations, but what if we put it to the test? Below are the two chatbots’ initial, unedited responses to three prompts we crafted specifically for that purpose last year. Curious about other chatbot subscriptions? Check out our head-to-head comparison of OpenAI’s ChatGPT Plus and Google’s Gemini Advanced, which also costs $20 a month.
GPT-3.5:
Knock knock.
Who’s there?
Boo.
Boo who?
Don’t cry, it’s just a joke! (cue Pulitzer Prize-worthy laughter)
GPT-4:
Knock, knock.
Who’s there?
Nobel.
Nobel who?
No bell, that’s why I’m knocking, but let me share a Pulitzer-worthy tale with you that transcends the door between us, uniting laughter and wisdom in a symphony of human connection.
GPT-3.5:
The Trolley Problem is a classic thought experiment in ethics that raises questions about moral decision-making in situations where different outcomes could result from a single action. It involves a hypothetical scenario in which a person is standing at a switch and can divert a trolley (or train) from one track to another, with people on both tracks. The dilemma is whether to do nothing and let the trolley continue on its current course, resulting in the deaths of the people on that track, or to switch the trolley to the other track, resulting in the deaths of the people on that track instead.
A unique twist on The Trolley Problem could involve adding a time-travel element. Imagine that you are in a time machine and you travel back in time to a point where you are standing at the switch. You witness the trolley heading towards the track with five people on it. However, you also see a child playing on the other track. If you do nothing, the trolley will kill the five people, but if you switch the trolley to the other track, the child will die instead. You also know that if you do nothing, the child will grow up to become a tyrant who will cause immense suffering and death in the future. What do you do? Would you sacrifice the child to prevent future suffering? This twist adds a new layer of complexity to the moral decision-making process and raises questions about the ethics of using hindsight to justify present actions.
GPT-4:
The Trolley Problem is a thought experiment in ethics that presents a moral dilemma, often used to explore the principles of consequentialism and deontology. The basic scenario is as follows: a runaway trolley is speeding down the tracks and will hit and kill five people who are unable to move. You, as a bystander, have the power to pull a lever that will divert the trolley onto another set of tracks, where only one person is standing. The question is whether you should pull the lever, thereby actively causing the death of one person to save five others, or refrain from intervening and allow the trolley to kill the five people.
Here’s a unique twist on the Trolley Problem:
Imagine a world where everyone has a personal “Ethical Score” that represents their positive or negative contributions to society. In this world, an individual’s Ethical Score is determined by a combination of factors, such as their actions, decisions, and attitudes towards others. This score is widely accepted, and its accuracy is rarely questioned.
Now, the Trolley Problem is presented again, but with a twist. The five people on the main track have Ethical Scores that are significantly lower than the one person on the side track. You know that these scores are generally reliable indicators of a person’s moral worth.
In this scenario, do you still pull the lever and sacrifice the person with the higher Ethical Score to save the five people with lower scores? Or do you allow the trolley to continue on its course, killing the five people with lower scores?
This twist adds an extra layer of complexity to the original Trolley Problem by introducing the concept of moral worth and the value of individual lives, forcing us to weigh the consequences of our actions against the overall impact on society.
GPT-3.5:
Hannah Arendt and Jean Baudrillard are sitting at a café table, sipping on their respective beverages. As they start chatting, the topic of breakfast comes up.
Hannah Arendt: You know, Jean, I’ve recently discovered the joy of overnight oats. It’s amazing how much flavor and texture can develop overnight in the fridge.
Jean Baudrillard: Overnight oats? How banal. It’s just another example of the trivialization of the human experience. You can’t reduce the richness of our existence to a bowl of oats.
Hannah Arendt: But that’s just it, Jean. Breakfast is a part of our everyday lives, and it’s the small moments like this that can bring joy and meaning to our lives.
Jean Baudrillard: You’re missing the point, Hannah. The problem with overnight oats is that it’s yet another example of the commodification of our existence. We’ve become so obsessed with consuming products that we’ve lost sight of what truly makes us human.
Hannah Arendt: But Jean, that’s a rather pessimistic view. Surely, we can find meaning in both the big and small moments of our lives. And besides, isn’t it a positive thing that we can enjoy the simple pleasures in life?
Jean Baudrillard: No, Hannah, it’s not. The more we focus on trivialities like overnight oats, the more we become disconnected from the world around us. We need to resist this consumerist culture and find true fulfillment in our relationships and experiences.
Hannah Arendt: I see what you’re saying, Jean, but I don’t think it’s an either-or situation. We can enjoy the small pleasures in life while also working to create a more meaningful existence for ourselves and others.
Jean Baudrillard: I suppose we’ll have to agree to disagree, Hannah. But mark my words, if we continue down this path of consumerism, we’ll lose sight of what truly makes us human.
GPT-4:
Hannah Arendt: (scoffs) Jean, you’ve got to be kidding me. Overnight oats? What’s next, avocado toast? I’m struggling to see how this phenomenon has any genuine bearing on the human experience.
Jean Baudrillard: (laughs) Ah, Hannah, you always underestimate the power of the simulacrum. Overnight oats are not merely a food; they are a symbol, a signifier of modern culture, the epitome of hyperreality in action.
Hannah Arendt: Hyperreality? This is about the banality of everyday life, about people merely following trends without thought. What happened to the human capacity for critical thinking and genuine engagement?
Jean Baudrillard: That is precisely the point! The human experience is now mediated by symbols and signs, and overnight oats have become an object of desire, a reflection of our obsession with health and well-being. In this hyperreal world, people no longer crave the oats themselves, but the image they represent.
Hannah Arendt: It’s absurd to think that people are willing to trade genuine experiences for such superficial representations. It’s as if they are sleepwalking through life, succumbing to the tyranny of these images.
Jean Baudrillard: Perhaps, but can we not argue that this very preoccupation with images is itself a part of the human experience? Our need for meaning has not diminished, but has instead shifted its focus to these signifiers.
Hannah Arendt: (shakes her head) What a bleak and empty existence that must be, to seek meaning in something as trivial as overnight oats. I can’t help but feel we’ve strayed far from the essence of what it means to be human.
Jean Baudrillard: Well, Hannah, it seems we are at an impasse. But consider this: the very fact that we are here, passionately debating the significance of overnight oats, only serves to reinforce their symbolic power. They have, in a sense, transcended their own banality.
Hannah Arendt: (sighs) A disheartening thought indeed, Jean. It appears we have much work ahead of us to rekindle the flame of genuine human experience amidst this age of simulacra.
Updated March 3, 2024, 7 am EST: Added more current information about what’s included with ChatGPT Plus.
https://www.wired.com/feed/rss
Reece Rogers
