The explosive rise of generative AI and how it will change everything
Industries
Trends

The explosive rise of generative AI and how it will change everything

While generative AI is still in its infancy, many people already see a very promising future for the technology, with lots of exciting potential applications that will significantly impact our daily lives in the years ahead.
  • What exactly is generative AI and what can it do?
  • Increasingly personalised marketing and sales
  • Providing faster, more empathic customer support experiences
  • Generative AI helps writers produce content more efficiently
  • No more tedious note-taking: generative AI tools automate the process
  • Writing code using generative AI tools ushers in a new era of coding
  • Generative AI will build you a blog, website, or online shop
  • Keeping customers happy, engaged, and loyal? There’s an AI for that
  • Product and fashion design with AI: next-level creativity
  • Creating a natural-sounding voice made possible by generative AI
  • AI-produced video within minutes: from short clips to full-length movies
  • Generative AI speeds up the pharmaceutical drug development process
  • Making quality game design cheaper, faster, and easier
  • Generative AI creates surprisingly authentic, never-before-heard music
  • The good, the bad, and the ugly
  • A future with generative AI… what will that look like?

Humans are very good at creating original work, and up until quite recently, we didn’t think that technology would be able to compete with this. But we are being proven wrong, as machines increasingly encroach on human territory and provoke conversations on how technology like artificial intelligence (AI) — and generative AI in particular — will impact the human experience. Generative AI uses deep learning and natural language processing (NLP) not only to push but completely obliterate the boundaries of what was considered possible only a few years ago. Generative AI is fast becoming creepily good at being ‘creative,’ and it seems as though every industry in which human professionals create original work is currently up for disruption: from graphic design to coding, gaming to marketing, sales to product design, law to architecture, content writing to social media, and any other creative industry you can think of. Changes are happening faster than the speed of light, and generative AI is increasingly moving into the mainstream. While — for now — we can leverage generative AI to assist us as an idea generator, springboard, or prompter, what will this technology mean for the future of the creative industry? One thing is for sure, for all its lofty promises and supposed potential, we will carefully need to consider the societal and ethical implications of this technology as well. 

Now, without further ado, let’s dive right in.

What exactly is generative AI and what can it do?

For the longest time, the question on everyone’s lips has been whether or not our creative jobs would eventually be taken over by technology as well. And up until recently, we genuinely believed that this would not be the case. Well, not soon, anyway. But with the explosive rise of generative AI in recent months, things seem to have taken quite a turn. Suddenly, it does seem possible for artificial intelligence to ‘be creative’ and interact more and more ‘naturally’. So, what exactly is this revolutionary technology that has taken the world by storm? 

Generative AI is a type of artificial intelligence that uses unsupervised deep learning algorithms and natural language processing (NLP) to create new content that is similar to (or rather, based on or derived from) existing content, such as text, code, audio, digital images, or video. Generative AI is taking assistive technology to an entirely new level, bringing incredibly powerful applications to non-technical users and significantly decreasing the time (and related costs) it takes humans to create this type of content. In short, generative AI will unlock faster, cheaper, and in some cases, better creations across many markets and industries. This will undoubtedly lead to significantly increasing productivity levels and economic value, and slowly but surely take over the work of writers, musicians, photographers, videographers, podcasters, meme-makers, designers, software developers, and so on. Generative AI can already design logos, do fundamental programming, produce music and videos for advertising, create stock photos, write blogs, emails, and recipes, explain scientific concepts, help build apps from scratch, create personalised therapy bots, and more.

“We are doing what Jasper and Copy.ai do but for video production. Videos are powerful — just imagine if marketers can send emails with talking human avatars instead of plain text”.

Josh Xu, cofounder and CEO of Movio

Increasingly personalised marketing and sales 

Instead of being limited by human creativity, generative AI can help people in sales and marketing generate new ideas using only a simple prompt. This is incredibly useful for those who want to build innovative and increasingly personalised experiences and engage with their audience in new ways — and within much shorter time spans. Generative AI can help identify customer preferences and create personalised captions, website pages, emails, and advertisements that are uniquely tailored to a specific customer. The technology can also assist with storytelling and generate highly personalised and engaging narratives instead of using predetermined, one-size-fits-all templates. Generative AI can also help generate more website traffic by identifying keywords and keyword phrases that will yield higher search engine rankings — and, ultimately, sales. Once the parameters have been set up and prompts have been entered, these smart systems require little to no human involvement, freeing up marketers for tasks like community building, social media moderation, strategic planning, creative ideation, and so on. 

The San Francisco-based startup SellScale uses generative AI to create more effective, natural-sounding, and personalised emails at scale. The generative AI  has been trained with high-performing outbound emails written by humans. SellScale then personalises those emails using data from more than 40 publicly available sources, such as articles, RSS feeds, and social media — but also from clients’ CRMs. The generative AI only keeps using the successful emails to further adapt, refine, and perfect its models with minimal involvement from sales development representatives. SellScale caters to education, Saas, healthcare, fintech, and consumer sectors. The technology integrates with tools like LinkedIn, Gmail, Zapier, Apollo, and Outreach, and works in close collaboration with sales teams. SellScale founder Ishan Sharma, who previously held a position at McKinsey’s Growth, Marketing, and Sales Service, explains: “We don’t want to be another tool to their dozens. Many competitors measure the value of their product on how much time salespeople spend inside their product. We measure value with how much salespeople don’t have to use SellScale or prospecting tools to write outbound”.

Another platform that’s becoming popular in the marketing sector is the one developed by Movio. The two-year-old startup created the platform to help marketers automatically create professional-grade videos with hyperrealistic, completely original text-to-speech ‘human’ avatars from scratch — and within minutes. It uses GAN machine learning frameworks and generative AI and features drag-and-drop interfaces and various templates. You can add a virtual representation of a human ‘spokesperson’ in the form of a fully customisable avatar — even in terms of its voice and emotional expressions — that you can instruct what to say via text input, with no video editing experience needed. The technology can be used to increase engagement through storytelling, video marketing campaigns, exhibitions, social media posts, product launches, and immersive experiences. Movio cofounder and CEO Josh Xu explains: “We are doing what Jasper and Copy.ai do, but for video production. Videos are powerful — just imagine if marketers can send emails with talking human avatars instead of plain text”.

Providing faster, more empathic customer support experiences 

Companies providing customer support services are facing record call volumes and struggling to manage the workload while waiting for outdated systems to provide the information needed to assist increasingly frustrated customers. To solve these problems and try to keep their customers happy, more and more companies are turning to generative AI solutions. Platforms like GPT, Bard, and Anthropic, for instance, are created with large, pre-trained language models that enable users to use text prompts to create unique content. This helps support agents to work more efficiently, create more personalised customer experiences, and resolve issues in a shorter time span. Generative AI enables the development of smarter, more ‘empathetic’ chatbots than ever before that are capable of understanding, anticipating, and responding to customer concerns and problems much more deeply and ‘naturally’. This is possible after they are trained with customer resolution data and case notes taken by support agents. Conversation patterns are then analysed, which enables the chatbots to identify trends and continuously improve the quality of support and information they provide. Generative AI can help supercharge productivity and deepen customer relationships. This will significantly alleviate the pressure human support agents are experiencing and free up time to get more involved in more complex and meaningful client interactions.

One example of generative AI for customer service is Yext Chat, a chatbot-building platform by online brand management and business information tech startup Yext. Yext Chat, an extension of the company’s Yext Search engine that was launched in 2019, combines various large language models from developers like OpenAI to enable incredibly helpful and coherent conversations in customer service support. To ensure information accuracy, Yext Chat links to its ‘Knowledge Graph’, as opposed to ChatGPT, which is trained with publicly available databases. The Knowledge Graph is an in-house database of company information and includes data about their locations, promotions, product launches, and even their employees. This data curation may have its limitations, but it does prevent Yext Chat from spouting biased or factually incorrect information, something that generative AI systems like ChatGPT and Bing Chat are known to have done. Yext Chat is continuously kept up to date and improves its answers so that it gets more and more in tune with how people interact with the chatbot. 

Yext Chat can easily be integrated into existing platforms and workspaces. A retail business could use it to assist customers with questions about their order status or to provide information on the company’s return policy. A clinic could use it to power a chatbot that provides patients with information on which medical professional to see and schedule their appointments. Yext COO and president Marc Ferrentino explains: “We believe it’s critical that every organisation start to understand what AI can do for them. Yext Chat is a transformative product that will provide every business with world-class conversational experiences that are safe, reliable, and easy to manage. Combining large language models with our Knowledge Graph unlocks a tremendous amount of potential and opportunity for our customers. Recent innovations brought conversational AI to consumers. Yext is bringing conversational AI to the enterprise”.

Generative AI helps writers produce content more efficiently

Writing engaging, high-quality content can be challenging and time-consuming, and generative AI can lend content writers a helping hand. The creative writing sector has been flooded with generative AI tools lately that all promise to generate ‘new’ and ‘original content’ that is virtually indistinguishable from human-written content — based on a prompt or some basic input — in significantly shorter time frames. The content produced by AI writing tools like Writesonic, Sudowrite, Longshot, GetGenie, Copysmith, Copy.ai, Jasper, and Writer is derived from millions of pieces of written text in articles, research studies, websites, books, and so on. AI writing tools can help writers come up with new content ideas or strategies, produce headlines and descriptions, write personalised email campaigns, rework existing content, and monitor how a certain piece of content performs in a marketing campaign. Based on analyses of customer preferences, purchase history, and behaviour, generative AI tools can generate content that is specifically tailored to the customer and include personalised subject lines and product recommendations. AI writing tools can significantly lessen the writer’s cognitive load and simplify the entire content-generation process. Generative AI should be seen as a helping hand for content writers and will require human editors to ensure the accuracy and appropriateness of the content.

The B2B-trained AI writing platform Writer, for instance, is specifically designed for the needs of enterprises and teams. Writer offers users the option to custom-train the language model on their own style guidelines and data to enable writing teams to generate articles, emails, and website texts that are in line with their brand’s guidelines, perform optimally, and achieve the desired results. The tool can carry out real-time internet crawls and make use — and sense — of various types of content, such as databases, spreadsheets, PDFs, audio, and even video material, in order to research, analyse, generate, and rework content for new purposes. Writer comes with over thirty original templates that can be used to create sales emails, social media posts, ads, research summaries, blog posts, and so on, or companies can create their own custom templates. Writer’s AI content generator, Ask Writer, can automatically answer questions, brainstorm ideas, generate drafts, and reference any website pages submitted by the user to produce even more relevant and accurate content. Writer offers a wide range of features, such as an intuitive drag-and-drop editor, customisable templates, and an AI-powered content assistant. Writer also provides analytics tools to help businesses track and measure the performance of their content. Additionally, Writer is a cost-effective solution that is scalable and can be used by businesses of any size. 

No more tedious note-taking: generative AI tools automate the process

Sales reps often complain about how tedious and time-consuming it is to take (digital) notes during meetings and sales calls, as it means having to record, replay, and sometimes even manually transcribe the conversations in order to organise follow-ups and plan further action and appointments. Generative AI tools can offer solutions for these problems by automating the process. The AI software can be programmed to recognise certain words and phrases and automatically generate a note based on the conversation. This means that the sales rep no longer needs to take manual (or digital) notes. The software can also replay what was said during the meeting or sales call, enabling the sales rep to quickly familiarise themselves with what was discussed and respond accordingly. The AI software can even transcribe the conversation in real time, enabling the rep to review and immediately take action on any important details.

One such note-writing tool is Notion AI. This ‘connected assistant’ helps you be more productive and more inventive in the workspace. Notion AI can help you read documents, create content, or take notes easier, faster, and more efficiently, and saves you time on basic text production and formatting. With special AI blocks like ‘summarise’, ‘action items’, and ‘custom content’, you can quickly generate precisely-tailored content. These blocks can even be incorporated into database templates to simplify the meeting process. Structuring ideas and getting concepts on paper to share with clients or colleagues can also be made more efficient with NotionAI, as it helps you rapidly progress from rudimentary notes to refined documents or presentations. All you need to do is create a bulleted list of key focus areas, performance requirements, sales targets, or achievements, and ask NotionAI to ‘formulate a paragraph summarising this material’ or ‘make this sound more professional’.

Writing code using generative AI tools ushers in a new era of coding 

By converting natural language prompts into code, generative AI tools significantly improve the efficiency of coding and provide those who may not have a coding background or much experience with opportunities to learn the coding process. Developers that use these tools are able to take on more challenging or expansive projects without having to spend hours of time coding manually. Generative AI coding tools also provide a significantly higher level of accuracy and consistency than manual coding and assure that the code follows a specific set of rules. This saves time and energy and enables developers to focus on more complex projects. Generative AI tools are ushering in a new era of coding, one that is more accessible, efficient, and accurate. With their help, coding is no longer a daunting task and frees developers up to explore their creativity and get involved in projects that were previously unimaginable.

One popular and innovative generative AI coding tool is the one made by Microsoft’s GitHub. GitHub’s programming tool, Copilot, helps developers with code suggestions to enable them to create code quickly and efficiently. It also helps them avoid common mistakes and typos that can — retrospectively — be time-consuming and difficult to fix. The tool can also quickly detect bugs before they become an issue, create user-friendly interfaces, and identify and suggest solutions to issues and problems that may arise during development. Copilot uses OpenAI’s large language model (LLM) to generate source code from textual descriptions and can perform a variety of tasks, ranging from auto-completing a line of code to writing full blocks of code. Since its launch, Copilot has integrated the feedback of millions of coders to optimise its model, significantly improving the quality of code suggestions and shortening the time needed for their delivery. According to GitHub’s most recent report, Copilot can help speed up coding by up to 55 per cent and can have huge benefits for enterprises. “With more accurate and responsive code suggestions, we’re seeing a higher acceptance rate for code suggestions. This means that developers using GitHub Copilot are staying in the flow and coding faster than before — and as a result —  are more productive and happy”, says Shuyin Zhao, GitHub’s senior director of product management. 

Generative AI will build you a blog, website or online shop 

Generative AI is a powerful tool for web app builders as well. It enables them to quickly and easily create complex programs without needing to code from scratch. With the help of generative AI, developers can use pre-trained models to automate certain development processes and create apps that are tailored to their specific needs. The technology is becoming increasingly popular among web app builders, with companies like Debuild and Wix providing various AI-based solutions to help make the process even faster and easier. 

The San Francisco-based startup Debuild has launched an AI-powered, low-code web application builder that makes creating and managing web applications simple and ‘blazingly fast’, enabling users to turn their idea into reality in minutes. Its graphical user interface allows you to visually assemble an interface and then deploy it in one click. The Debuild platform can automatically generate React components and SQL code, eliminating the need for manual coding. It offers a range of features, such as a drag-and-drop editor, customisable designs, and integrated e-commerce and analytics tools. It also offers a wide range of templates and themes with which you can easily customise web applications to look exactly how you want. Debuild’s powerful code editor enables users to make changes to their web applications at the drop of a hat, and its versatile collaboration tools make collaboration with other developers easy as pie. Debuild’s founders write: “At Debuild, our mission is to build an autonomous system that can create software at the level of the world’s most skilled engineers. By giving every person the ability to instantly create any software tool they need, we can radically alter the trajectory of human progress”.

Another popular platform that uses AI-based tools — and is known for being a site builder that offers stunning templates — is Wix. Wix also has an integrated e-commerce platform and a comprehensive app marketplace and features powerful analytics tools that enable developers to detect errors and optimise performance. Wix’s new artificial design ADI (artificial design intelligence) makes use of a machine learning algorithm to create beautiful and engaging web applications. The ADI is ideal for beginners who quickly want to build a blog, website, or online shop without needing any coding knowledge. ADI is more user-friendly than the Wix Editor and even provides an interactive tour to familiarise users with the software. Users who prefer more creative control and more options, such as various design elements, more than 900 templates, and around 250 apps, can switch to the Wix Editor at any point. 

Keeping customers happy, engaged, and loyal? There’s an AI for that

Businesses are continuously searching for ways to improve their customer experience and increase revenue. Many strategies that can achieve this, however, are often time-consuming and expensive. This is where generative AI can assist as well. It promises to optimise the customer experience and improve personalisation by providing automated solutions. For instance, generative AI can help automate communication with customers. Natural language generation technology can automatically generate personalised chat and email responses, ensuring that clients quickly receive relevant and accurate information. Generative AI can also help organisations expand their client base by providing a continuous stream of targeted leads, identifying upselling and cross-selling opportunities, and informing customers of new products or services. The technology can also be used to process customer input and contextual data to generate valuable insights, which enables increasingly personalised product recommendations and can even help prevent customers from leaving. This technology is far more accurate than anything previously available and offers intelligent, creative solutions that are tailored to an individual customer or to a specific target group. Generative AI enables businesses to gain value from their customer relationships and increase customer engagement with precisely tailored products, which results in increased customer engagement and satisfaction, as well as improved brand loyalty.

“Rather than starting with a drawing, the designer starts with the parameters required for the end product. The system then generates thousands of potential solutions in just a few hours and identifies the top few options that best fit the requirements”.

Peter Champneys, Autodesk

Product and fashion design with AI: next-level creativity

Generative AI also has the potential to revolutionise fashion and product design. In fact, in the years ahead, a vast majority of product design companies will increasingly rely on AI as it offers the ability to explore thousands of potential design solutions. Peter Champneys, a research engineer at Autodesk, explains: “Rather than starting with a drawing, the designer starts with the parameters required for the end product. The system then generates thousands of potential solutions in just a few hours and identifies the top few options that best fit the requirements. The human contribution is to then evaluate the outputs and determine the best performing, most cost-effective and visually attractive solution”. Generative AI tools that can be used for fashion design, such as Midjourney, use advanced algorithms to analyse fashion trends and create unique designs. The tools are capable of understanding a variety of fashion design elements — such as colours, patterns, textures, and silhouettes — and learning from user feedback. This enables them to adapt their designs to better fit the user’s preferences. Generative AI tools can be used to quickly create a variety of different fashion designs, which can then be used for further customisation and refinement by a human designer, helping fashion designers save time, increase efficiency, and create more unique and innovative designs.

In his latest collection, Metamorphosis, AI artist, fashion designer, and keynote speaker Baris Gencel showcases the vast scope of what AI art can lead to. Gencel explains that a concept he refers to as ‘inconceivable fabric’ — AI-generated, experimental materials — can really push the boundaries of traditional fashion design. “By incorporating these unique fabrics into their collections, designers can create truly one-of-a-kind pieces that stand out in a sea of sameness”, Gencel says. “Whether it’s a dress made from a strange plastic-like material mirroring plastic bottles or a jacket woven from plant-textured bamboo fibres, these unconventional materials add excitement and innovation to fashion design that helps it stand out in an increasingly crowded market”. Gencel has also been approached by many about creating digital wearables and collectables for communities in the metaverse — a promising market that is developing quickly and will increasingly blur the line between physical and digital reality.

Creating a natural-sounding voice made possible by generative AI

Generative AI in speech synthesis enables machines to generate a natural-sounding voice from text input. The technology is based on deep learning algorithms that are trained on vast amounts of data, which enables the machines to understand the nuances of human speech and generate natural-sounding audio. AI in speech synthesis is used in applications like virtual assistants and chatbots, but also in assistive technology that helps people with visual impairments ‘read’ text content. Two examples of generative AI in speech synthesis are Google’s AudioLM and Voice Design.

Google’s AudioLM is a deep learning-based voice synthesis technology that uses techniques from large language models to create natural-sounding audio, complete spoken sentences, and even continue piano music. It is able to generate audio from parts of speech like nouns and verbs, as well as from complex sentences, and can produce audio in various languages, dialects, and accents. Maximilian Schreiner, managing editor at The Decoder, explains: “Google AudioLM was trained with 60,000 hours of English speech, and another variant was trained with 40,000 hours of piano music. Both models use semantic and acoustic tokens and can continue speech and music of previously unheard speakers and pieces after their training”. Voice Design, also from Google, is a platform that uses generative AI in speech synthesis to produce natural-sounding voices for applications. The platform is used to create voices for virtual assistants, chatbots, and other applications and uses deep learning algorithms to ensure its voices sound more natural than those produced by traditional text-to-speech methods. Like AudioLM, it can also generate voices in multiple languages, dialects, and accents. A popular branch of speech synthesis is audio-visual speech synthesis or multimodal speech synthesis, which makes use of an animated face that is tightly synchronised with the synthesised speech. 

AI-produced video within minutes: from short clips to full-length movies

Yes, ambitious levels of filmmaking are now also accessible to the masses. Generative AI can help you create all kinds of video content, ranging from short clips to entire movies. It does this by using generation techniques that use input data like blogs, articles, images, and music. Image generators create the visual content, while text generators are used to produce storyboards and scripts, and music generators are used to compose the soundtrack. Generative AI tools for video content also offer a wide range of creative options to customise videos, from colour correction and photo-realistic rendering to motion tracking and object recognition. One popular AI video generator is the one created by Synthesia, a startup founded by a team of entrepreneurs and AI researchers. The Synthesia tool, which is used by the BBC, Nike, Accenture, Google, Reuters, and many other big names, enables you to create any type of quality video content you can think of — within minutes and without studios, microphones, or cameras. Synthesia offers more than 100 diverse preset AI avatars (or can create custom-made ones), various templates, voiceovers in 120 languages (including closed captions), a media library, a screen recorder, and more. Each language also has a variety of tones to choose from. For instance, you can have your video sound serious, professional, empathetic, warm, factual, natural, easygoing, lifelike, and more. Whether you’re creating training or marketing videos, you can select your message’s tone — just as you would if you recorded it yourself.  Synthesia can be used to produce professional product marketing videos, how-to posts for social media, educational videos, or any other type of video content, and eliminates the need to travel to filming locations, hire expensive professional actors, or the use of complex and costly video equipment.

“Early foundation models like ChatGPT focus on the ability of generative AI to augment creative work, but by 2025, we expect more than 30 per cent — up from zero today — of new drugs and materials to be systematically discovered using generative AI techniques. And that is just one of numerous industry use cases”.

Brian Burke, research VP for Technology Innovation at Gartner

Generative AI speeds up the pharmaceutical drug development process

Did you know that, on average, it takes approximately ten years and billions of dollars to develop a new pharmaceutical drug? Advances in deep generative models are about to change all of this, however. By utilising AI’s ability to generate and analyse numerous potential drug candidates, researchers can quickly identify the best options and focus on them. This significantly streamlines the drug development process and leads to great time and cost savings, which enables the production of more efficient drugs at a quicker pace. Generative AI can also be used to predict how potential drugs might behave in the body, significantly decreasing the need for painstaking lab work. Brian Burke, research VP for Technology Innovation at Gartner, says: “Early foundation models like ChatGPT focus on the ability of generative AI to augment creative work, but by 2025, we expect more than 30 per cent — up from zero today — of new drugs and materials to be systematically discovered using generative AI techniques. And that is just one of numerous industry use cases”.

Sean McClain, founder and CEO of Absci, a Washington-based firm that uses AI to search through billions of potential drug designs, explains: “We’re seeing an uptick in activity and investment because increasing automation in the pharmaceutical industry has started to produce enough chemical and biological data to train good machine-learning models. We’re going to see a huge transformation in this industry over the next five years”. Experiments on tissues and cells, as well as human clinical trials, are still critical steps in the process, and the ultimate validation still needs to take place in the lab. These are the most time-consuming and costly aspects of drug development. Generative AI is, however, already saving a lot of time by automating many steps that humans used to do manually. And while it could still take a number of years before the first pharmaceuticals designed with generative AI will become available, the technology will most certainly transform every aspect of the pharmaceutical industry in the years to come.

Making quality game design cheaper, faster, and easier 

A computer game can cost millions to produce — from writing the storyline and creating the graphics and soundtracks to developing the media and marketing content for it. The gaming industry is, however, increasingly making use of generative AI to make it cheaper, faster, and easier to produce quality games. Basically used as a ‘turbocharger’, generative AI enables human game designers to create unique new levels, generate dialogues for non-player characters, write interesting storylines, and design virtual environments — like jungles, deserts, cities, or planets — for players to explore. With generative AI, all of this can now be done in much shorter time spans and with much lower costs involved. Platforms like Promethean can be used to create virtual worlds, while Charisma, Convai, and Inworld are popular options for creating non-player characters (NPCs). AI will not only fast-track the development of games, but it will also enable the creation of more personalised and dynamic games that adapt to gamers’ preferences. Some current examples are games like Hidden Door and Dungeon, which already enable you to design a custom avatar based on a few typed sentences. And in the future, this technology will enable designers to create virtual worlds from scratch much faster and with even greater ease.

The popular gaming platform Roblox is currently testing an AI code-writing tool that might significantly speed up the process of creating and changing in-game objects. It may, in fact, enable players themselves to design avatars, terrain, and buildings, adapt how these look and behave, and even give them interactive properties — all by merely typing instructions in natural language instead of using code. Roblox CTO Daniel Sturman, for instance, demonstrated that typing ‘reflective, purple foil, crushed pattern’ into a chat window changed the way a sports car looked in the game. He even showed how typing ‘Blink the headlines every time the user presses B’ made the sports car do exactly that. Roblox explains that the AI it uses to create the code is powered by capabilities from outside sources, as well as in-house technology. “The approach holds promise for Roblox because so many of the games on the platform are made by individuals or small teams. We have everything on our platform, from studios down to 12-year-olds who have had an incredible idea come out of a summer camp”, says Sturman.

“While parts of the music production process will be replaced with generative platforms like Aimi, I do not see a future where art goes away”.

Edward Balassanian, CEO of Aimi

Generative AI creates surprisingly authentic, never-before-heard music

Generative AI is also increasingly used to produce original music compositions. It works by feeding data into a deep-learning algorithm to create a generative model that can compose new music. The model is trained on vast datasets of music, which enables it to understand and draw from the complexities of musical styles, rhythms, and tones. It then creates new, never-before-heard music that is surprisingly authentic, with a range of different styles, tempos, and instruments. In 2016, Google’s Magenta composed the first-ever AI song, and the technology has been improving at a record pace ever since. The most significant development Magenta has predicted in terms of the impact of generative AI tools on music production is the creation of completely new music genres.

A very popular generative music creation tool is Aimi. It can create an infinite number of immersive electronic musical compositions based on music provided by artists. What’s different about this tool is that it offers an innovative new way of experiencing music as it evolves and adapts to the feedback of the listener. After being presented with a library of source materials, the AI analyses, organises, and reassembles these to compose a generative musical experience. The tool thoroughly restructures and manipulates the original audio in real time according to Aimi’s algorithms. Then the artist tweaks the results, leading to an endlessly unfolding composition that does not repeat itself. The listener can use the thumbs up or down to provide feedback to the software. The tool then makes adjustments in the composition and shapes the music to the preferences of the listener. Aimi provides generic musical experiences that are categorised by mood, but it also provides artist-branded experiences that are created based on audio provided by well-known DJs and producers across various musical genres. Aimi CEO Edward Balassanian says: “Our platform allows artists to create generative programs that effectively mix, master, and produce music in real time, replacing much of the tedious process of hand-assembling music in studios. Aimi seeks to fundamentally change the way the art form is created, consumed, and monetised”. He continues: “While parts of the music production process will be replaced with generative platforms like Aimi, I do not see a future where art goes away”.

“Generative AI is a very convincing liar. It is able to generate rather impressive-looking ‘original’ content, but as it is solely based on training data, we need to realise that the accuracy and originality of this content can leave much to be desired. So, no matter how good technology seems to be performing, the importance of due diligence — by humans — should not be underestimated”.

Richard van Hooijdonk, trendwatcher & futurist

The good, the bad, and the ugly

As we’ve seen from the many examples in this article, generative AI’s main advantage — and this is predominantly from a business perspective — is the fact that it reduces our reliance on human involvement in the creation of content. It is able to generate (new) text, images, video, and music much faster and cheaper than humans can. It can also help us understand complex systems, analyse data, generate important insights, and improve decision-making processes faster than ever before — which, in turn, vastly improves efficiency and productivity. Another important advantage of generative AI is that it enables us to generate increasingly personalised content faster and with much greater ease and that it opens up new possibilities for generating unique solutions and creating new avenues for innovation.

The introduction of generative AI has also led to heated debates about this disruptive technology’s impact on society. To start off, for all its lofty promises and hyped capabilities, generative AI certainly does not always produce accurate or good results. Its algorithms often base their output on either false or biased information — which can lead to or perpetuate harmful stereotypes. They are also known for making up citations and fabricating URLs. This leads to questionable or sub-standard quality content at best and the creation of offensive and even dangerous content at worst. In the words of renowned futurist and trendwatcher Richard van Hooijdonk: generative AI is a very convincing liar. It is able to generate rather impressive-looking ‘original’ content, but as it is solely based on training data, we need to realise that the accuracy and originality of this content can leave much to be desired. So, no matter how good technology seems to be performing, the importance of due diligence — by humans — should not be underestimated”. 

Furthermore, with organisations always on the lookout for faster, more cost-effective ways to produce content, it is only a matter of time before many occupations will be rendered obsolete, and many professionals’ ability to earn a decent living will be significantly reduced. Professor of art Carson Grubaugh, for instance, predicts: “Large parts of the creative workforce, including commercial artists working in entertainment, video games, advertising, and publishing, could lose their jobs because of generative AI models”. Aside from this impact — as if that isn’t worrisome enough — there’s the potential for abuse of the technology. One example is students using generative AI to ‘write’ academic papers. Another is the stories about increasingly prevalent copyright infringement issues and the privacy and personal data violations that have flooded the internet in recent weeks. What’s more, generative AI is an unfamiliar and complex technology, and many don’t know how it works or how to implement it, which is counterproductive to its applications and advantages. Many experts, even, don’t know exactly how the technology works. The introduction of this technology is also bound to lead to a society that’s divided into those who’ll benefit and those who’ll increasingly find themselves on the losing end of the scale. Then there’s the lack of control. Once a trained generative AI model is let loose, there’s no telling exactly what type of content it might generate, and in the event it’s not pretty: how to stop it. 

A future with generative AI… what will that look like?

While generative AI is still in its infancy, many people already see a very promising future for the technology, with lots of exciting potential applications that will significantly impact various industries — and our daily lives — in the years ahead.  More and more developers will introduce new and exciting tools and platforms that promise even more applications, higher speeds, and better efficiency. Going forward, it is highly likely that generative AI will also be increasingly integrated with technologies like natural language processing and computer vision in order to develop even more versatile and powerful capabilities. Experts predict that content like articles, website content, images, videos, and program coding will quickly become predominantly AI-produced, which will simultaneously make deceptive, inaccurate or offensive content more complex to not only distinguish but prevent. This will, in turn, lead to the development of new generative AI tools that can combat the spread of misleading or unethical content.

Going forward, there is a strong need to carefully govern the use of these generative AI models and their impact on society, a task not only reserved for the private sector but for policymakers and civil society as well. Only when robust guidelines and checks are in place will we be able to achieve — and benefit from — the many advantages generative AI technologies undoubtedly have to offer.

I asked an AI writer to create a few paragraphs about the short-term future with generative AI. This is what it said: 

“In five years, a future ruled by generative AI will be a reality. This new technology will be integrated into every aspect of our lives, from our homes and workplaces to our transportation and entertainment. AI will be used to optimise processes and drive efficiency through predictive analytics, allowing us to achieve our goals faster and more efficiently. We will be able to access personalised services tailored to our individual needs and preferences, delivered at a speed and precision that only AI can provide. 

In the workplace, generative AI-enabled robots will take on many of the mundane tasks, freeing up humans to focus on more creative and innovative pursuits. We will be able to collaborate with AI to generate new ideas and solutions, furthering our progress as a society. AI will also enable us to develop new technologies and products with faster speeds, increased accuracy, and advanced capabilities. 

In our homes, AI will help us automate mundane tasks, like grocery shopping and cleaning, so that we can spend more time with our families and friends. AI will also be integrated into our healthcare system, helping to diagnose diseases, develop personalised treatments, and improve overall wellbeing. In five years, the world will be a very different place, but one thing will remain the same: humans will continue to be the driving force behind the advancement of technology and society. We will still be the ones creating, innovating, and pushing the boundaries of what is possible”.