Category: Artificial Inteligence
-
Originality
AI has moved the idea of derivative work into the headlines. After all, originality makes the greats great, or does it?
“Great Artist Steal” – Steve Jobs
I mean, think of all the original work we love—music, movies, tv, and art that genuinely shaped us.
Art is about the creation of something, and something sometimes comes from something. That something original becomes in itself a new art form.
The best part is sometimes a cover, a recreation of what we love and believe – but it also comes from sampling what we like and love but just the right taste.
But don’t go too far. A sample is good, but too much could be a copy.
But let’s not forget. Real artists steal, right? And art is about creating something not just new but from what we learn or acquire. I learned to write from reading and typing – Now is the time for all good men to come to the aid of your country.
I mean, when do we, as humans, get caught up with the past? Not us. Nostalgia never gets the best of us – but even if it did, we would make it better, and better, and better, and better, and better, and better, and better, and better, right?
So how dare AI copy all these original ideas we have!
I think we need to protect everything and keep the web or anyone from knowing things like a world-famous authentic New Orleans Red Beans recipe.
So is generative AI any different than our own derivative work? I mean, we can’t seem to figure that out – and sometimes, even some folks can’t find the fun in a little joke. Ideas come from so many people I know, so much I’ve read, and so much I want to know.
But maybe DALL-E and Open-AI find it kind of funny?
Originality is a really complex topic, and the impact of AI on jobs is scary. Shouldn’t we want the same inputs that allow us to create to be the things we feed the AI we use? Is that original work generated from the same tweets and arrives I read as derivative as my own?
In the end, we can all agree that copying can create some of the best art – or maybe we’re all just here for the ride.
-
Is AI the end of Search Engine Optimization?
In testing the new crop of AI ChatBots, I noticed a consistent and fascinating trend, hunting for search results really sucks, and I want someone to tell me what I want. I want the AI bot to give me the answer. Not Bing or Google result pages or yelp listings. Tell me what I want to know. Here is a recent scenario: I’m heading to NYC to do college tours with my daughter and staying at a hotel in New York near NYU. The Sunday before, I plan to do brunch with two friends. One is a pescatarian, and the other is vegetarian – I eat practically anything and everything. We want the place to have a casual environment so we can hang out at a table or couch with drinks after and be within walking distance of my hotel.
I hit Google and manifested this as the search term “Brunch spots near NYU with vegan options.”
Google’s first set of results overlays some great-sounding places on a map. Now, I’m looking for locations or jumping at whatever name sounds great. The ratings are all 4+ stars or higher, so that’s not necessarily useful.
If I continue to scroll, I’m met with SEO (Search Engine Optimization) SPAM pages of the ten best lists that are AI-generated puff pieces with no actual human curation or Yelp’s search SEO spam pages that often have nothing relevant to my search.
Yelp when clicked shoes a search on the topic with the upper top 1/3 of the page full of ads and sponsored posts that are not relevant to the search. Sadly, Yelp did not have a list of places near NYU, so I was met with a list of restaurants in Manhattan.
I get a list of options in the remaining results – some good, some not. Genuine articles from actual New Yorkers, local bloggers, or local neighborhood papers are nonexistent.
So after getting frustrated and feeling defeated, I decided to ask ChatGPT.
This is the full transcript:
There are many great brunch options in New York City, but here are a few recommendations that meet your criteria:
1 Cafe Mogador – This East Village cafe is known for its Mediterranean-inspired brunch dishes, including Shakshuka and Moroccan couscous. They have plenty of vegetarian and pescatarian options, as well as meat dishes for you. The atmosphere is casual and relaxed, with indoor and outdoor seating options.
2 Bluestone Lane – This Australian-inspired cafe has several locations throughout the city, including one near NYU. They offer a variety of brunch items, including avocado toast, breakfast bowls, and a smoked salmon bagel. They also have a selection of vegetarian and pescatarian options, and their atmosphere is cozy and welcoming.
3 Jack’s Wife Freda – This cozy cafe in Greenwich Village has a Mediterranean-inspired menu with plenty of vegetarian and pescatarian options. They’re known for their shakshuka and their grilled haloumi cheese sandwich. The atmosphere is casual and inviting, with indoor and outdoor seating.
All three of these options are within walking distance of NYU, have vegetarian and pescatarian options, and a casual atmosphere that’s perfect for hanging out after brunch.
From here, I chatted with ChatGPT and added additional thoughts and filters, like wanting a place that served mimosas and one with traditional American breakfast options. I picked a spot, made reservations, and was done.
The experience made me realize that hunting for search results and the growth of SEO SPAM has turned search into something that no longer produces results. Are the results from ChatGPT genuine? No, but it feels more authentic and direct to my request. It gave me an answer. This content may not tap into alternative New York magazines or the hip new blogger, but it feels definitive.
What is funny is that Google could have gotten here quicker. It did get here faster. Many protested the company’s urge to skip showing results and provide direct answers or snippets. If you remember, Google said you need a definition; why link to 15 different dictionary sites if you can answer the question? This generated tons of fear among small businesses and digital marketers as it quickly cut off a powerful funnel for leads. Worse was the fear that the logic for deciding this would be primarily money driven and hidden from the larger population.
Folks that is what ChatGPT does and exactly what we’re craving. Would you like Siri, Alexa, or Google Assistant to tell you where to go for dinner and not give you a list of 8 million possibilities? I would. Choice may not be what we are looking for after all. Right? Right… -
Is ChatGPT a Big Deal?
For the last few weeks (maybe months), OpenAI and several of its products have been in the press. If I knew nothing, this was the first time anyone had ever experienced AI or witnessed the immense amount of AI-based products we have consumed as customers for years.
It’s not as if demos have shown AI ordering food for us. It can’t look at an image and determine the type of plant it is. Can’t it translate words for us in real time? Can’t AI write articles or create music?
All of the things ChatGPT and recent AI innovations do are old news. We’ve been riding this wave since the video Humans Need Not Apply was recorded in 2014.
What has changed is access to this tech. Before OpenAI, using AI required complicated training models and hidden developer-centric tools. You needed product teams to attach APIs and write code. OpenAI is a reflection not of innovation but of what happens when the technology available to anyone becomes easy to access. OpenAI ripped down walls and created a simple interface to showcase the ongoing development for years.
If LL Cool J were to sum it up, he might say, “Don’t call it a [revolution]. I’ve been here for years.”
-
It Wasn’t Me
Arstechnica has an amazing piece on the dystopian possibilities of AI images and deep fakes. As the article notes – deepfakes have been a reality for years, but AI takes what was a skill and makes the process so simple that anyone can channel its uses for good (or primarily bad). It opens up a world where the art of disruption is limited to our ability to capture a picture in our imagination and transcribe it as words.
I can remember joking at how insane a song like Shaggy’s It Wasn’t me, when in reality, maybe Shaggy was a sage describing our pending deepfake future. For those thinking the computers won’t fool me try out this quiz. Times are early, and fake images are becoming harder and harder to pick out.
-
DALL-E
DALL-E and the world of generated images have captured the attention and imaginations of many. When I first watched the video, humans need not apply; close to 8 years ago, the technologies felt possible but still distant. DALL-E alerts you with a flashing red light to how far we’ve come over a relatively short period.
DALL-E is an AI-based system that generates realistic images from a string of text. What makes it uncanny is its understanding of the text and its ability to apply context and do all of this across different artistic stylings. Using the tool is addictive; if you have not, I suggest you create a free account and give it a whirl.
Our CEO Todd has also turned me on to many other AI tools, like jasper.ai that allow you to generate blog post articles with a simple topic or description. While they may miss the depth and meat many expect in a well-crafted post, it is a shockingly great starter (and better than some content on the Internet).
What I find fascinating in the new AI space are the same copyright issues we struggle to answer around ownership, especially when referencing prior art. For example, one can sample music and use it to make a new song, but we have defined a line that determines when a new song is unique and, in other cases, when the sampling requires the artist to pay royalties to the previous artist.
In the case of tools like DALL-E, the prior art is exactly how you train a machine to create something new or unique. You give it as many samples of images and artwork as possible and provide it with metadata to describe each piece of work. It allows you to ask it to generate an image of a dog in the style of Van Gough.
Is this a case of a new unique piece of art? To what extent is it based on the prior works that AI used to create this new piece of work? Are the uses of training sources any different than me asking a human to do the same thing? If one profits from the work, who should receive the royalty? The engineer who developed the AI? The company who created the AI? The license holder who typed a string of text to generate this new work of art? Or maybe the AI itself?