Automation has come a long way, leading now to its most advanced and buzzworthy state, artificial intelligence. AI refers to technologies that learn from training data and experience to perform tasks that would otherwise require human intelligence. When applied to a PM’s job, AI enables “lights-out” project management, in which software handles the project from quoting to invoicing without the need for human interaction. Over the last few years, CSA Research has observed a growing number of LSPs that operate with business models that are fully or partially automation-driven.
AI exists at the end of a continuum running from simple trigger-based automation, through more robust rule-driven expert systems to big data-driven applications that learn from observing projects:

• Rule-based systems deliver basic automation for configured workflows. LSPs preload price sheets, workflows, and stakeholders in the system and enable it to execute the project plan based on clear rules. For example, when a given client sends a project in Spanish, the software automatically runs the files against that client’s translation memory (TM), produces the quote, and sends the pre-translated file to a pre-assigned translator for that language pair. However, if the system encounters an undefined situation, such as a new customer, or new language for a client, the pre-configured workflow stops and calls for human intervention.


• Expert systems. Intelligent systems go a step beyond pre-loading data for each scenario. Instead they apply complex rules to quote rates and turnaround times, choose workflows, and select vendors based on project specifications. In this touchless environment, human intervention occurs when the system flags a need for it. For example, the software may discover a shortage of vendor options that can handle the work in the assigned turnaround time.


• AI with machine learning. As systems move beyond rules, they learn on their own by analyzing data in the absence of explicit direction. For example, an AI-driven system can predict timelines based on actual translator performance for specific types of content. It flags the odds of a translation passing a preset quality threshold based on analysis of events such as whether the linguist opened a provided glossary. It can also draw previously unanticipated conclusions from escalations to improve the handling of similar cases in the future.

The range of automation can vary greatly, but developers tend to focus on project intake, project management, vendor management, and project post-processing. 

Even the most sophisticated lights-out systems we’ve observed retain some human-centered elements. For example, if LSPs can handle projects from A to Z with their AI, then vendor managers can focus on the relationship with suppliers, and account managers can invest more effort in developing client strategies.
AI in project management is bound to trigger a complete revolution in the long-held and prevalent beliefs of the language services industry. While executives may be tempted to resist the push for automation out of fear of the unknown or lack of technology expertise, automating PM processes becomes even more urgent for LSPs than adding AI via neural machine translation (NMT). CSA Research contends that the benefits it delivers in eliminating unnecessary manual touches will allow companies to re-deploy its human assets to more valuable tasks. As the technology improves, we recommend that all providers review their operations to learn where they could take best advantage of AI.
Of course, AI makes the language industry anxious. Even after LSPs switch to heavily automated business models, there will still be people involved. Some LSPs will take advantage of the changes, others won’t, this is no different than what we’ve been seeing for the last 30 years. There is risk, and work to be done, but the sky is not falling.
Our latest research, “Will AI Eliminate the Need for Project Managers?,” analyzes the state of automation at LSPs based on responses to our annual global market survey and provide 12 steps to prepare for AI. For additional data and insight on automation adoption patterns in LSP production teams, refer to “The State of Project Management at LSPs.”