How to navigate the impact of generative AI

What are the implications of generative artificial intelligence and large language models for the print industry?

2023 was the year that generative artificial intelligence (AI) emerged. 100 million people now use market leader ChatGPT – at the time, it was the fastest-growing application in history. Like other technological leaps forward, there is no stopping its integration into many aspects of our work and personal lives.

Generative AI marks a significant step forward in technology, enabling machines to produce human-like text, images, and responses. But AI is not a conscious being (however surprisingly ‘real’ it sometimes seems).  It learns from billions of words and phrases gathered from billions of sources, predicting what words are likely to come next to create an often uncanny, human-sounding presence in text.

This can lead to lightning-fast, human-sounding responses, but can also feature factual error, copyright issues and has caused concern for existing workforces, who fear their jobs could be made redundant. However, we should try and treat scary news stories about AI with caution and see it as just another tool that can be used in your workplace. One that if used correctly and with the right guardrails in place can only benefit your business.

Smaller, more agile firms are well placed to make the most of AI and use it to save time on mundane tasks and free up staff to work on more profitable projects

Large language models, exemplified by OpenAI’s GPT-3, are advanced AI systems honed on expansive internet datasets. Through unsupervised learning, they navigate millions upon millions of phrases, extracting patterns and nuances from diverse text sources like articles and books. With billions or trillions of parameters, models like GPT-3 showcase contextual understanding, and excel in tasks from translation to creative writing. In particular smaller firms, which are more agile, are well placed to make the most of AI and use it to save time on mundane tasks and free up staff to work on more profitable projects.

Opportunities and pitfalls

For example, AI can be used in market research, offering nuanced insights through text analysis. In marketing, it streamlines content creation and personalises messaging at the click of a mouse. In time it will become a convenient, omnipresent tool for anyone working with language or data, much like a pocket calculator.

Generative AI tools can boost programmer productivity by automating repetitive tasks such as writing code, generating test data, automating unit testing, troubleshooting code errors, and drafting documentation.

Customer service can benefit too. LLMs enhance customer interactions by deploying chatbots, virtual assistants and automated communication systems. This leads to smoother customer service and less manual effort. A report from Accenture showed that LLMs could be useful in tackling 70% of customer service interactions that are not straightforward and could benefit from a conversational, powerful and intelligent bot.

In creative teams, AI can extrapolate new images from existing ones, generate new photos based on a person’s face or place an image of a product in a new location. Or customers can describe, in words, what design they would like on a t-shirt and AI can supply options for customers to choose from – much as DALL-E generates artwork by interpreting verbal inputs.

In some applications, AI can be outsourced to customers themselves when personalising their own products. At Antigo Designer, customers can employ AI for facial detection in photos, showcasing the tangible personalisation capabilities. Beyond aesthetics, AI-driven imposition facilitates more efficient print runs and saves waste.

For printers, this improved productivity is a obvious benefit, particularly with automated content creation tools that expedite marketing material production

The benefits across the company are clear. A new study on the impact of generative AI on highly skilled workers found that it could improve a worker’s performance by as much as 40% compared with workers who don’t use it. The implications for workforces are still being worked through, but it seems likely that instead of causing mass unemployment, AI will still require human cognitive inputs to ensure outputs are correct. Employees will be able to cope with greater volumes of work or spend more time on human-facing tasks such as customer support, sales or research and development.

For printers, this improved productivity is a obvious benefit, particularly with automated content creation tools that expedite marketing material production. However, occasional errors are inherent in the technology, ranging from misinterpretations to accuracy issues. Additionally, navigating the challenge of distinguishing fact from fiction in content generation poses a real concern, especially in contexts requiring factual accuracy. Copyright issues have also arisen because of ChatGPT’s training on publicly available sources such as news sites.

Next steps with AI

In the print industry, the practical applications of generative AI focus on automation, workflows, and robotics. AI optimisation of workflows enhances precision and speed in tasks, while the integration of robotics anticipates workflow changes, addressing potential issues proactively for continuous productivity. Labour costs have risen around the world, and as manufacturing flows back to once-defunct manufacturing bases in the US and Europe in the drive to ‘onshore’ industries, AI can help printers find much-needed savings.

The impact on marketing and market research translates into practical advantages for businesses: personalisation of marketing materials and informed decision-making through trend analysis for example. Moreover, the time-saving capabilities of AI enhance overall productivity, redirecting focus towards high-value tasks, and the integration of robots trained by AI contributes to operational efficiency.

Printers should stay informed and optimistic about AI applications and innovations, and ready to seize the opportunities it brings

Generative AI’s predictive capabilities could enhance just-in-time manufacturing in the print industry. Aligning production processes with anticipated market demands becomes an operational strategy, minimising excess inventory and reducing waste for more efficient production. For example, businesses could predict fashion trends, supply chain problems, or commodity prices by analysing massive amounts of language data. Sentiment analysis will allow businesses to track their online reputation and quickly fix any problems that arise.

Printers should stay informed and optimistic about AI applications and innovations, and ready to seize the opportunities it brings. They should also be clear-eyed about the disruptive potential of AI, particularly in the profound challenges to workforces and their changing roles. In addition, the compute power required to scale up the use of AI should not conflict with existing sustainability measures. If you don't have an AI policy in place already, now is the time to start working on it.