中文Since the public launch of ChatGPT in November 2022, artificial intelligence has dominated technology headlines. But for the printing and media industry, AI is not a new arrival — it has been quietly optimizing production processes for years. The difference now is that generative AI is moving from the pressroom to the design studio, creating new opportunities and new questions for print service providers.
Weak AI on the Production Floor
The printing industry has been using what Knud Wassermann, editor-in-chief of Graphische Revue, calls “weak AI” — more accurately, machine learning — for specific targeted tasks across the production chain for many years. Image data acquisition and recognition systems control quality across all process steps, and when deviations from reference data are detected, software automatically compensates on the fly without manual intervention. This applies equally to analogue and digital printing machines.
The results are measurable in Overall Equipment Effectiveness (OEE), a business metric that measures the percentage of actual productive manufacturing time. According to Heidelberg, the average OEE for offset printing machines is currently slightly over 30 percent. Individual printers that rely heavily on automation and optimization are already achieving peak values of more than 60 percent, depending on their order mix. Cloud-based solutions that integrate machines into broader data ecosystems are improving order sequencing, setup times, spoilage rates, and process stability. Preventive maintenance concepts driven by machine learning keep availability high by predicting failures before they occur.
Generative AI Arrives in the Studio
The more visible disruption is happening in creative workflows. Adobe’s Firefly platform, demonstrated with generative AI models integrated into Photoshop, Illustrator, and Express, promises that “everyone can transform their own words into creative ideas, regardless of their personal knowledge.” For print service providers, this opens up possibilities for providing increased graphic design support to customers — generating layouts, creating variations, extending images, and producing marketing collateral with reduced dependence on specialist design resources.
But generative AI also raises questions that the industry has only begun to address. What is the data basis on which AI models are built? Where does the training data come from, and how is diversity represented? Who holds the copyright — the company that developed the AI solution or the user who prompts it? In professional environments, the responsibility for verifying and validating AI-generated information clearly lies with the user, but the practical mechanisms for exercising that responsibility in high-volume production workflows are still evolving.
The Content Authenticity Question
Adobe is attempting to address one dimension of this challenge through the Content Authenticity Initiative (CAI), which aims to create a global standard for trusted attribution of digital content. The goal is to embed a universal “tag” in file information that would enable image creators to exclude their content from being used to train AI image generators, while AI-generated content would be appropriately marked. The tag would be linked to the content wherever it is used, published, or saved.
For print service providers navigating the intersection of traditional production expertise and emerging AI capabilities, the path forward requires balancing the efficiency gains of automation — both on press and in prepress — with the judgment and quality control that only experienced professionals can provide. The tools are becoming more powerful. Using them wisely, and knowing when human judgment must override algorithmic output, will define the successful print businesses of the AI era.
The sustainability dimension of AI also deserves attention. Generative AI models consume substantial computational resources during both training and inference, and data centers supporting AI workloads have significant energy and water footprints. For an industry that is increasingly required to report on its environmental performance — driven by regulations like the EU’s Corporate Sustainability Reporting Directive (CSRD) and customer demands for supply chain carbon accounting — the energy consumption of AI tools used in prepress, design, and workflow automation may need to be factored into sustainability reporting.
The question of copyright in AI-generated content is particularly acute for the printing industry, which occupies the intersection of creative work and industrial production. If a designer uses an AI tool to generate a layout that incorporates elements trained on copyrighted images, who bears the liability if the original rights holder objects — the designer, the print service provider, or the AI platform? These questions are being litigated in courts around the world, and definitive answers may be years away. In the meantime, print service providers who use AI tools in their workflows should implement policies that require human review of AI-generated content and maintain records of the tools and data sources used in production.
Perhaps the most constructive way to think about AI in the printing industry is not as a replacement for human expertise but as an amplifier of it. Machine learning on press makes skilled operators more effective by handling routine adjustments automatically and flagging exceptions for human attention. Generative AI in the studio makes designers faster by handling repetitive layout variations and letting them focus on creative direction. In neither case does the technology eliminate the need for human judgment — it raises the stakes for getting that judgment right, because the speed at which AI-assisted workflows produce output means that errors can propagate faster than ever before. The printing businesses that thrive in the AI era will be those that pair powerful tools with disciplined quality processes, using automation to handle what can be automated while maintaining rigorous human oversight of what cannot.
Source: drupa blog, “AI in the media and printing industry,” guest article by Knud Wassermann, Editor in Chief of Graphische Revue.
