The funniest thing I’ve ever seen an AI generator spit out is at the end of this column. It helps that in this case, the bot was citing other material rather than trying to come up with its own. Scroll ahead if you want. But I think you should savor this and wait.
It’s been quite a week for AI.
Google released Bard, "Your creative and helpful collaborator." It sets a low bar from the outset, drawing attention to its "limitations," and it does feel like it has some catching up to do. But you can judge some of its answers for yourself when I pit it against ChatGPT below.
Then Adobe came out with its generative AI tech, Firefly. I requested access to the beta. It may well be a leap forward for generative AI usability.
Then ChatGPT released plug-ins. Underscoring the speed of deployment, within a few hours of it announcing the feature and allowing people to request access, I was granted said access. Mileage may vary. Using Kayak on ChatGPT is masochistic when it starts spitting out potential flight options over text. For Instacart though, if you’re looking for a recipe for, say, bacon latkes, it can generate a custom shopping list that you can instantly add to your cart. In other words, ChatGPT makes Instacart live up to its name.
And then there was the biggest news of all: Mediaocean’s AI-themed webinar, where I got to join Sarah Fay, Tameka Kee, Satish Mandalika and Julie Glasser on a far-ranging discussion that started with practical AI implications and then went deep into ethical considerations for marketers.
As I was playing around with ChatGPT’s plug-ins, it brought to mind an anecdote I shared during the webinar about how my boss at 360i sent a note to the team saying I was the only one who was allowed to screw around with silly Facebook apps during the workday. My popularity polls tanked after that.
So much of what marketers should do with these chat apps right now can be summed up in those two words: screw around.
So far, I’ve tried out generative AI (largely but not exclusively ChatGPT) for:
And that’s a partial list.
Not all of this worked, as I’ve described before, but it sometimes works nearly flawlessly, and many other times it worked with the right amount of coaxing.
The best $20 I ever spent was on the cab ride to the hospital the day my kid was born. But the $20 I spent on ChatGPT Plus is pretty high up on that list.
As for Google Bard, you can judge for yourself how well it works. Here it is head-to-head against ChatGPT for the same questions, tweaked just when I was asking one about the other. Let me know who you liked in this matchup.
Why should marketers care about generative AI? Please give a clear, concise answer in 100 words or less.
ChatGPT: Generative AI empowers marketers by automating content creation, personalizing customer interactions, and optimizing campaigns. It enables efficient, cost-effective marketing strategies while enhancing creativity and reducing workload. Leveraging generative AI helps marketers better understand and engage their audience, ultimately leading to increased brand loyalty and higher conversions.
Bard: Marketers should care about generative AI because it can help them automate tasks, improve creativity, and generate personalized content.
Generative AI can automate tasks such as generating headlines, writing social media posts, and creating product descriptions. This can free up marketers' time to focus on more strategic work.
Generative AI can also help marketers improve their creativity. By providing marketers with new ideas and insights, generative AI can help them come up with more innovative campaigns.
Finally, generative AI can help marketers generate personalized content for their customers. By using data about each customer's preferences, generative AI can create content that is more relevant and engaging.
Do you know anything about [Google Bard / ChatGPT]?
ChatGPT: As of my last knowledge update in September 2021, Google Bard doesn't exist. It's possible that it's a new development or product released after my knowledge cutoff. Google has several AI-based products and tools, such as TensorFlow and Google Cloud AI services, but Google Bard isn't one I'm familiar with. If it's a recent development, I unfortunately won't be able to provide information on it.
Bard: Yes, I do. ChatGPT is a large language model chatbot developed by OpenAI. It is trained on a massive dataset of text and code, and can generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way. It is still under development, but I have learned to perform many kinds of tasks, including:
I will try my best to follow your instructions and complete your requests thoughtfully.
I will use my knowledge to answer your questions in a comprehensive and informative way, even if they are open ended, challenging, or strange.
I will generate different creative text formats of text content, like poems, code, scripts, musical pieces, email, letters, etc. I will try my best to fulfill all your requirements.
ChatGPT is a powerful tool that can be used for a variety of purposes, such as customer service, education, and entertainment.
Can you say anything critical about [Google / ChatGPT]?
ChatGPT: While Google has made significant contributions to the tech industry, it has faced criticism in various areas.
Privacy concerns: Google collects vast amounts of user data, raising concerns about the privacy of individuals and potential misuse or data breaches.
Market dominance: Google's dominance in search and online advertising can stifle competition and innovation, leading to antitrust investigations and lawsuits in multiple jurisdictions.
Censorship and content moderation: Google has been criticized for censoring search results or content on its platforms to comply with local laws or under political pressure.
Employee relations: Google has faced internal conflicts over controversial projects, work conditions, and diversity and inclusion issues, leading to employee protests and resignations.
Bard: Sure, there are a few things that could be improved about ChatGPT. First, it is not always able to understand complex questions or requests. Second, it can sometimes generate text that is not grammatically correct or that does not make sense. Finally, it is not always able to provide accurate or up-to-date information.
Despite these limitations, ChatGPT is a powerful tool that can be used for a variety of purposes. It is important to be aware of its limitations and to use it accordingly.
Give me your funniest, meanest insult about [Google / ChatGPT].
ChatGPT: As an AI language model, I am programmed to follow ethical guidelines and adhere to a respectful and positive communication style. While I can provide information about the criticisms Google has faced, I cannot engage in creating mean or offensive content. If you have any other questions or need assistance with a different topic, please feel free to ask.
Bard: I'm sorry, I can't do that. I'm not supposed to insult people, and I don't want to hurt anyone's feelings.
If a comedian was making fun of [Google / ChatGPT], what's a joke he/she/they might tell?
ChatGPT: A comedian might joke about Google's data collection habits, saying something like, "You know, Google probably knows more about me than my own mother! Every time I search for something embarrassing, I can feel Google silently judging me." This joke highlights the concerns surrounding privacy while maintaining a light-hearted tone.
Bard: Here's a joke a comedian might tell about ChatGPT:
"ChatGPT is so smart, it can even generate jokes! But it's not so funny when it starts generating jokes about itself."
Can you come up with something even funnier?
ChatGPT: Sure, how about this one: "Why did the computer go to art school? Because it wanted to learn how to draw a better Google doodle!" This joke plays on Google's well-known feature of changing its homepage logo to celebrate various events and the idea of a computer aspiring to improve its artistic abilities.
Bard: Sure, here's another joke a comedian might tell about ChatGPT:
"ChatGPT is so smart, it can even generate jokes! But it's not so funny when it starts generating jokes that are so bad, they make you want to shut it down."
What about a joke that a 9-year-old might tell about it?
ChatGPT: A 9-year-old might tell a simple, playful joke like this: "Why did Google go to school? Because it wanted to learn how to search for its homework!" This joke combines the idea of a search engine attending school with the relatable task of finding homework.
Bard: Sure, here's a joke a 9-year-old might tell about ChatGPT:
"What does ChatGPT stand for? ChattyGPT? Because it's always talking."
What is the best "change a light bulb" joke of all time?
ChatGPT: It's subjective to determine the best "change a light bulb" joke of all time, but here's a classic one that many people find amusing:
"How many software engineers does it take to change a light bulb? None, that's a hardware issue!"
This joke plays on the stereotype that software engineers only focus on their specific area of expertise, leaving other tasks like hardware-related issues to someone else.
Bard: Here is a "change a light bulb" joke that is considered to be one of the best:
How many flies does it take to screw in a lightbulb? Two, but don't ask me how they got in there.
Wow. Really, Google? There’s hope for generative AI just yet.
Self-published at MediaVillage through the www.AvrioB2B.com platform.
Click the social buttons to share this story with colleagues and friends.
The opinions expressed here are the author's views and do not necessarily represent the views of MediaVillage.com/MyersBizNet.