By Nathalie Brähler
Black Friday 2019.
Six people, four women and two men still in the office. Coffee cups and empty soda cans on the table. They’re visibly tired after 12 hours of working non- stop. One group member is standing next to a white board, leading a creative session while the rest of the group gives input and votes on ideas. After reaching consensus, the outcome is scribbled on post-its and moved to a job board, the whiteboard is wiped and another team member starts a new session. Every now and then, one of them walks over to a dot matrix printer in the corner that is spitting out a long piece of paper while steadily typing sentences, adding to the 1 meter high pile of paper already covering the floor.
Six hours earlier the team checked the printed sentences with laughter, while reading them out loud. Now, the atmosphere seems a bit more grim and the new sentences are checked in silence.
For that same Black Friday, content marketing assistant Persado has been writing headlines and short content that should generate a high conversion rate for clothing retailer Gap: to create messages that feel personalised to Gap’s customers across channels, to make them click on the sales link. With Black Friday online sales hitting high, it is important for the retailer strike the right cord to meet ever-evolving consumer demands.
Meanwhile, content creator Mark is doing the same task as Persdao, for a different client, knowing the amount of clicks could define the next step in his career: the higher the click through rate, the more he will be appreciated by the team. If Mark is playing his content cards right, this promotional Friday could skyrocket the yearly sales figures. Every now and then he checks his Gmail inbox and he shares his input as fast as possible with the team, after which he’s back to content writing again.
On a different continent, in a different time zone, Billie is handling customer service questions for online retailer Bol.com. Especially during ‘Black Friday week’, where over 25 orders per second are placed, it’s essential he gives all customers the information they need, navigating them through the ordering process.
In all these examples, artificial intelligence was part of the work force. Billie is Bol.com’s virtual assistance, or chatbot. During Black Friday, he is able to do the work of 225 FTE,1 handling over 70% of all the customer service questions successfully. He learned from the data of over 50.000 conversations with customers and only when Billie cannot answer the question, he passes it over to a human colleague. During Valentine’s Day, Billie is assisted by bot Cupido on Facebook. Cupido suggests a last-minute Valentine gift to customers based on some personality questions, reaching a substantial higher conversion rate than the human customer service desk during Valentine.2
Mark is a human, using the automated smart compose function in Gmail, to speed up the mail process. Via Google’s machine learning models it makes text suggestions and weaves in Mark’s past writing to personalise prompts.
Persado is an artificial intelligence robot that writes marketing copy which is incredibly effective, reaching as high as a 450 percent lift in click-through rates on ads3, while humans as Mark would reach a 50-200 percent rate.
The dot matrix printer in the corner of the office where the team is working late, is an AI that writes funny copy with awkward, snark and dark humor that would please the customers of the game Cards Against Humanity. The game creators taught a computer how to write typical Cards Against Humanity cards and they put it to the test on Black Friday. During 16 hours, the writers battled the powerful writing algorithm4 to see who could write the most popular new pack of cards. If the writers would win, they’d get a $5,000 holiday bonus. If the A.I. would win, the writers would be fired5. Fans could follow the battle via a muted live cam and they could up- or downvote cards on the human and AI side. Eventually, they could buy either the AI deck, “30 cards actually written by a machine learning algorithm trained on our brainstorming documents” or the Human deck, “30 beautiful cards about the human condition straight from the hearts of our human writers.” Despite this somewhat biased advertising for the humans, the human writers sold only 2% more packs, which made the Cards Against Humanity company react dryly: “So their jobs will be replaced by automation later instead of right now.”
The writers must have foreseen this conclusion because they took the last hour of their long battle to brainstorm about “Other jobs we’d be good at” jotting down jobs as cult leader, producer and Hollywood movie star on the whiteboard. Surprisingly, the team went mostly for traditional positions that are, because of AI, under the same changes as their writing jobs.
At Creative Business University of Applied Sciences Utrecht, we claim to train our students for jobs that don’t exist yet. In order to do so, we need to look beyond the emotional discours around “the threat of AI for our jobs”.
There are different ways to look at the role of AI in the Creative Business in general. AI can be seen as ‘Augmented Intelligence’ where it is an assistant to humans. It augments personalised interactions6 with customers, co-workers, readers7 and suppliers, it can augment knowledge, being a treasure trove of data for humans to work with. To gain insights and to help identify and anticipate on- going, changing trends, making sure, for instance, that there’s the right amount of stock in periods of high demand8, or to offer a journalist the latest content or stories about a certain topic or person9. It can augment the creative execution in the process —simultaneously creating multiple versions while maintaining version control and achieving mass personalisation.10 Here, we see the AI entering the creative field, as an ‘Augmented Creativity’ tool.
Automated content writers such as Persado and automated journalism AI’s such as Wordsmith are successful examples of Augmented Creativity, that are used by big names in the industry. The AI’s take away the typical ‘donkey work’, the mundane tasks that are usually not high on the list of creative employees. Those mundane tasks were usually passed on to interns, so they could learn step by step without doing much harm if they would make a mistake. Now that the AI takes over, doing the tasks in seconds, without explanation and with often a flawless result, we are aware that internships are slowly but steadily changing. In companies that rely on AI for process-driven tasks, internships will resemble a normal job right from the start which can be more fascinating for the intern but it also requires a junior-level or sometimes even mid-level attitude. That’s why we train our students from year one onwards, on professional behaviour, about the ever changing technology -to avoid tech shyness- and the changing workforce worldwide. They need to be confident around chatbots11 and AI tools and they should have some experience with it before they enter their work placement.
But what if AI’s are not longer augments creativity but creates, composes itself, or even creates actors? Every year, the examples of a “creative AI” are getting more spectacular: IBM Watson wrote an entire tv commercial script for Lexus, advertising network Dentsu launched AICO, a very successful AI copywriter, Christie’s sold an artwork created by an AI for $432,500 and digitally de-aged12 or re-born humans13 are starring in box office movies or on podia.
Creativity no longer seems to belong uniquely to humans. Company Creative AI says we’re trending towards unprecedented access to creative tools (photo editors, music studios, etc) as well as greater opportunity for collaboration (social online platforms) which will lead to “the escalation of creativity”; or “a world where creativity is highly accessible and anyone can write at the level of Shakespeare, compose music on par with Bach, or paint in the style of Van Gogh”. This trending will have an effect on intellectual property (IP), business models, the value of creative work and the role of the audience and fans. In order to understand this, our Creative Business students work with creative AI tools in year four and assess these tools on various aspects, ranging from creative value to ethics and inclusivity. They learn about commercial AI tools such as Scriptbook14 which not only predicts box office success for film studio’s, it can also determine whether or not a film will pass the Bechdel Test, a bare minimum baseline for representation of women in media, or if the film will include a diverse cast of characters. We teach them to think critically about what seems like a good AI: scripts will need to specify a character’s race and whitewashing can occur later on in the process.
“It’s easy for AI to come up with something novel just randomly. But it’s very hard to come up with something that is novel and unexpected and useful” says John Smith, Manager of Multimedia and Vision at IBM Research. Maybe that’s the essence of what we need to teach students to prepare them for jobs that do not exist yet: to orchestrate all the data-output of the AI’s in such a way it becomes something novel, unexpected and useful.
—
Q’s run in talktotransformer:
Artist: Can I make money with AI?
Algorithm: Yeah you can, but it’s going to be a chore to code that AI to do anything but do all the repetitive tasks.
Artist: How can I make money with AI?
Can you make money by hosting an AI competition or even by providing AI services?
Artist: How will AI evolve art?
Answer: I think it will evolve into something similar to a spirit world. With this, we have to consider the user and AI’s value systems. AI can be very individualistic and selfish, whereas human beings have value systems and desires that draw them to a particular brand and piece of art, so creating them together is a great idea. As human beings develop technology, AI will evolve to match that, just as humans have a hunger to learn, and AI has an insatiable curiosity and desire to create more life. (>>checked in Grammarly and it’s 100% original)
How will AI evolve creative business?
A variety of technical advances from machine learning, artificial neural networks, big data, and deep learning are placing additional stress on the fields of art, science, technology, art, and music. We could begin by recognizing their importance and creating a collective push to create a more intelligent, collaborative, flexible, responsive, creative, and inclusive society.
What does civilization need?
Ultimately, it will need all the more our capacity for creative self-expression, infinite potential for new scientific discoveries, expansive knowledge, and a global community of knowledge-workers. We must have a creative and innovative culture to replace the old, stagnant one.
What are successful examples of AI in creative business?
For most successful examples, AI has to look similar to human judgment – not just understand it, but get it right more often. For instance, I give you a list of celebrities who are worth several billion dollars each. These are two of the people in the world who are worth most of their net worth just in brand names. They’re really good at branding – it’s really hard to move the needle with a brand name that’s worth as much as the CEO of Apple or the CEO of BP.
What are examples of AI in creative business?
Take your sight, for example. You need some kind of translation from a human language to human vision, say visual impression. That is how we can make sense of visual impressions in the world around us.
The same applies in creative business, like ecommerce. We need something that represents images in our minds, and we need to figure out the structure. If an image is simple, which is the opposite of image mnemonics, how do we know what it’s about? The same goes with marketing. How do you differentiate an ecommerce site from a newspaper page? Can we leave the screen by looking at the images we are consuming and search out the parts that relate? Or maybe we can take
What are some challenges that are disrupting creative industries?
Software is replacing human interactions. Computers now do one-fifth of all the work in the United States, at jobs that were once dominated by people. Over the next decade, every hour of every day we will live in a digital world that will make it possible for software to do everything from drive cars to make purchases to, yes, sit in our living rooms and participate in a lecture in our living rooms. It will enable us to work anywhere we go, to share ideas and collaborate with anyone and everyone, and to find and meet people from all over the world in a fraction of the time and cost of having to set up and run any kind of meeting, conference, or event.
What are some challenges that are disrupting creative industries?
“My biggest challenges are workflow. There are all kinds of ways that people are trying to make my life easier, to find ways that things are consistent. But what do I do to deal with all that?”
Margaret Lee, founder of MindBodyGreen, asks how we can collectively address the challenges that will shape our creative careers. “Can we do it in a way that puts work before personal stuff? Can we do it in a way that goes beyond just working with friends? Should we have to be smart instead of caring? We are up against a tidal wave of technological advances and the fact that we are losing touch with our principles, our time, and our creativity.”
One of the most common uses of artificial intelligence for marketing is to recognize patterns in customer behavior. The beauty of this sort of AI is that, although it requires some training (to narrow down your vision), it can then be easily applied across a vast volume of data and turned into highly effective marketing.
1. Huub Vermeulen, managing director Bol.com in his speech during Webwinkel Vakdagen, The Netherlands, January 2019.
2. D. Drost-deRidder, manager Service Innovation at Bol.com, during Chatbot conference, 2018.
3. https://www.persado.com/press-releases/jpmorgan-chase-announces-five-year-deal- with-persado-for-ai-powered-marketing-capabilities/
4. The AI is a neural network, borrowed from the open source GPT-2 model created by AI research company OpenAI and trained specifically to write Cards Against Humanity (CAH) cards. The GPT-2 model was already trained on over 40,000 books worth of internet text to ensure it can reliably predict and fill out the next word or punctuation mark in a sentence with realistic effect. CAH trained it further on the text of 44,000 cards.
CAH: “We stopped the training once it could “consistently” produce cards matching the grammar and tone of the game. We did this so it didn’t draw too much from already- written cards while still leveraging as much cultural information as possible.” After that, CAH ran a straightforward filtering algorithm to ensure it could select AI-generated cards that matched its standard format and to avoid ones that were too similar to existing cards.
5. If this would be really true is hard to tell but Cards Against Humanity claimed it on their special Black Friday Humans vs AI website: https:// www.cardsagainsthumanityaichallenge.com/
6. A study by Gartner predicts that by 2021, around 15 percent of all customer service interactions globally will be managed by AI.
7. The Perspective API tool organises reader’s comments interactively so that viewers can quickly separate the “toxic” comments from the “more illuminating” comments.
8. There’s a rise of startups in this field, such as iris.ai and klydo.ai help companies gather insights faster, filtering down to what’s really relevant through AI.
Picasso Labs and Automated Insights are the more established companies in this field, used by the Creative Business.
9. Juicer is a data extraction tool that gathers data from multiple news sources. It aggregates and extracts news articles, assigns semantic tags to the stories and organises them into categories.
10. Examples are: Adobe Sensei, assisting designers in video and image production, Albert, a full AI marketing platform automating media buying and targeting and Lobster, searching and licensing of social media imagery.
11. E.g. the course Advanced Marketing challenges students to create an explanatory chatbot as an assignment and it introduces them to neuro-marketing tools and intelligent influencer marketing platforms.
12. Samual L Jackson in Captain Marvel, Robert de Niro in The Irish Man and Will Smith in Gemini: examples of actors being de-aged with the help of AI tools.
13. Tupac Shakur was brought back to life as a hologram, performing at Coachella in 2012, opera singer Maria Callas was brought back to life in a hologram tour in 2018 and this year, actor James Dean will star in a new movie, 64 after his death.
14. Scriptbook claims that its algorithm is three times better at predicting box office success than human readers. The company also asserts it could have prevented 22 of Sony Pictures biggest box office flops over the past three years, which would have saved the production company millions of dollars.
Scriptbook’sAI system can predict an MPAA rating (R, PG-13, etc.), detect who the characters are and what emotions they express, and also predict a screenplay’s target audience.
Scriptbook says that their system could be used to aid human readers, not replace them.
