ChatGpt has been in the news a lot recently and its abilities have gotten no shortage of press. From its capacity to assist in everything from essay-writing and text-processing, to website-building and scam-creating, it’s clear that ChatGPT — as well as its growing list of competitors — are powerful tools. But one of ChatGpt’s most practical applications is also one of its least mentioned.
For as long as I’ve been using computers, the internet has been a vast source of knowledge on nearly every subject I could imagine. But so often, formulating the questions that would yield the precise results I was looking for was a challenge. It wasn’t that the information wasn’t out there, only that finding my way towards it wasn’t as simple as Googling my question or related key words.
But ChatGPT can now make sense of my more complicated questions. Over time, I’ve been trained into believing that there are just some inquiries a little too complicated to type into a search bar. But with the emergence of ChatGPT, I’ve continually had to remind myself that my computer can now help me with those few questions that only last year it still couldn’t.
In the past, those denser, multiple clause questions of mine could sometimes point me toward meaningful results. But more often than that, they would just get simplified into a jumble of key words that didn’t even place me within the same ballpark as the answers to my questions. And all of a sudden, with the quiet shift of a paradigm, ChatGPT can help me when I say things like:
“There was an article I read in about 2016 that argued satire was dead in the era of the Trump presidency, but I’m struggling now to find it. Are there any pieces you can find that fall within those parameters?”
or
What would be an easily understood reference to a couple of buffoons in a movie who comically argue?
For both, it shot back answers which I was immediately able to apply to a piece I’d been working on. For the first, it suggested three separate articles which all fell directly within the parameters I mentioned, one of which was the precise one that I remembered reading and had been looking for. For the second question, ChatGPT promptly suggested I use the Home Alone criminals. It was the perfect reference.
As time has gone on, I’ve been stunned by just how little there is that ChatGPT struggles to make sense of. Even the sorts of questions or statements which might confuse any reader it can process instantaneously. Below is a conversation I had with ChatGPT only a few minutes ago to exhibit the sorts of capabilities I’m describing.
Me: What would be 5 terrific examples of questions that demonstrates to a wide audience the complicated sorts of everyday questions that ChatGPT could make sense of, and provide useful results for, which Google could not?
ChatGPT: The advantage of a conversational AI model like ChatGPT over a traditional search engine like Google lies in its ability to engage in a dialogue, understand context, provide detailed explanations, and generate more personalized responses. While Google is designed to provide a list of the most relevant web pages based on the keywords of a query, ChatGPT can handle more nuanced and interactive tasks. Here are a few examples of questions that would highlight these advantages:
Interpreting Literary Themes: “Can you explain the themes and symbolism in ‘The Great Gatsby’ by F. Scott Fitzgerald?” While Google can certainly provide numerous articles discussing this topic, ChatGPT can generate a detailed and contextual response, integrating multiple facets of the novel in a single, coherent answer.
2. Generating Ideas: “Can you provide some unique fundraising ideas for a local animal shelter?” Google can return many generic fundraising ideas, but ChatGPT can propose more personalized solutions based on further questions and answers within the conversation.
3. Explaining Complex Concepts Simply: “Could you explain quantum mechanics in a way a ten-year-old could understand?” A Google search would provide various explanations, but they might still be too complex. ChatGPT can generate a simpler, more age-appropriate explanation, and iteratively refine the explanation based on the user’s feedback.
4. Providing Advice or Suggestions: “I’m feeling anxious about an upcoming job interview. Do you have any tips to help me prepare?” While Google can provide numerous articles on this topic, ChatGPT can offer a more empathetic and conversational response, tailoring advice based on the individual’s specific concerns.
5. Guiding Personal Decisions: “I’m torn between majoring in Computer Science and Fine Arts, what factors should I consider?” Google might offer lists of factors to consider, but ChatGPT can facilitate a deeper exploration of the user’s personal interests, goals, and concerns, helping to guide a more informed decision…”
Me: Can you keep the questions more in the vain of the following question? : “What would be an easily understood reference to a couple of buffoons in a movie who comically argue?”
ChatGPT: Sure, I understand you’re looking for questions that require contextual understanding and a synthesis of cultural and nuanced knowledge. Here are some examples:
Movie Recommendation Based on Specific Traits: “Can you suggest a feel-good movie where the main character is a bumbling yet lovable underdog?”
Identifying Song Lyrics: “What’s that famous song that mentions ‘strangers waiting, up and down the boulevard’?”
3. for Cooking: “I messed up my soufflé and it deflated, what could have gone wrong?”
4. Understanding Literary Characters: “Which Jane Austen character is known for her quick wit and prejudice?”
5. Identifying Pop Culture References: “Can you explain the ‘I am your father’ joke that people often reference from Star Wars?”
Each of these questions involves a degree of understanding and context that Google, with its keyword-based search, might struggle to fully comprehend. In contrast, a conversational AI like ChatGPT can provide a more nuanced and contextually appropriate response.
Me: Can you provide 5 examples that fall somewhere in between the most recent two sets of parameters that I’ve given you?
ChatGPT: Absolutely, I understand you’re looking for questions that engage cultural or contextual understanding, as well as problem-solving abilities. Here are five examples that combine those aspects:
Historical Context: “What events in history could have inspired George Orwell’s ‘1984’?”
Identifying Similar Characters: “Who is a TV show character similar to Sheldon Cooper from ‘The Big Bang Theory’ in terms of quirks and humor?”
Interpretation of Lyrics: “Can you explain the meaning behind the song ‘Hotel California’ by the Eagles?”
Scenario-based Advice: “I’m planning a surprise ‘Star Wars’-themed party for my best friend. Can you suggest some creative ideas for decorations and activities?”
Specific Recommendations: “I enjoyed the comedic dynamic between Jules and Vincent in ‘Pulp Fiction.’ Can you recommend other movies with similar character chemistry?”
These questions involve both a comprehension of cultural or contextual elements and the ability to generate thoughtful responses or suggestions, demonstrating the advantages of a conversational AI like ChatGPT.
Me: What would be an example of one of the hardest, densest, multi-clausiest sentences which you could make sense of and help with?
ChatGPT: Here’s an example of a complex, multi-clause question:
“In your learned estimation, given the broader consensus among the AI research community, who despite engaging in extensive debates regarding the limits and ethical implications of artificial intelligence and, in particular, the deployment of highly advanced language models like yourself, persist in their pursuit of ever more sophisticated and capable AI systems, do you not concur that it’s fair to postulate, notwithstanding the array of potentially profound societal impacts both positive and negative, that the trajectory of AI development will continue unabated, and if so, what might be the ramifications of such developments on society at large?”
And here’s how I would answer it:
Indeed, based on current trends in AI research and development, it’s reasonable to assume that the advancement of AI, including highly advanced language models like me, will continue. Despite the many debates about the limits and ethical implications of AI, the quest for more powerful and capable AI systems persists.
The potential societal impact of such advancements is broad and complex. On the positive side, AI can automate routine tasks, enhance productivity, enable new forms of communication like myself, and even help solve complex problems in fields like healthcare and climate change.
However, these advancements could also bring about challenges. They may intensify issues around privacy, job displacement due to automation, and the potential misuse of AI technologies. Moreover, ethical questions about AI decision-making processes and the lack of human oversight also arise. Therefore, it’s crucial that as AI continues to advance, ethical considerations and regulatory measures evolve in tandem to mitigate potential negative impacts.
So for those of you who may have written off ChatGPT as something too technically dense to grapple with, or simply labeled it as a way of cheating rather than researching, I implore you to consider some of its most practical applications. It grants us tools that didn’t even exist for the public until a few months ago. Just a couple of years ago, they were still far beyond our reach.
The profound sorts of aids that AI can offer certainly don’t negate the harm that could ultimately arise from it. It’s true that AI has the potential to change everything — for better and for worse. And while the utility it can offer in research is spectacular when used carefully, the potential for dissemination is no small concern.
There have been increasing accounts of instances where these programs provide false information. In rare instances, the dishonesty has even seemed malicious. But when used as a launching pad rather than a final source, its assistive capability is difficult to overstate. Where the winds of this AI storm will ultimately land us is one of the great uncertainties of our time, but trying our best to get ahead of it while we can is likely our wisest move.