Author: Jesper Lehrmann
Product Director, LanguageWire
LinkedIn
Security issues with machine translation
Author: Jesper Lehrmann
Product Director, LanguageWire
LinkedIn
Machine translation security.
Are you considering machine translation? Many people are. Free translation tools with increasingly sophisticated machine translation engines are becoming more and more popular – and that goes for business translations as well. Companies understand that to reach a large, global audience, you need to speak more languages. We’ve discussed the multilingual content boom and how it plays out on the web, with LanguageWire and as part of the latest translation trends.
If the need for multilingual content is always increasing, it seems machine translation, with its cheap cost and fast turnaround, perfectly meets this need. In fact, according to Common Sense Advisory, post-editing of machine translation is the third largest language service in 2017 only behind translation and on-site interpreting. But with the rise of data breaches in popular businesses and the start of GDPR, you have to ask: how secure is machine translation?
Depending on which tools you use, machine translation can be either a great way to reduce security risks or a security disaster.
According to a 2017 study from IBM, although the global average cost of a data breach is down 10% to $3.62 million, companies are having larger breaches. The average size of data breaches increased 1.8% to more than 24,000 records. Losing data is expensive.
Using machine translation to automate translation of sensitive content, such as confidential documents, might help you improve security and avoid these costs. The more automated the translation process, the fewer touchpoints and shorter transit time in which data can be stolen. This also means fewer people are likely to gain access to data that might be top secret. In fact, simple errors by employees, like opening attachments or clicking on links from cyber attackers, are some of the most common ways businesses lose valuable data.
Source: IBM X-Force Threat Intelligence Index 2018
Machine translation engines control access to the content and have automated processes, resulting in fewer people-related vulnerabilities. But this type of protected machine translation requires a dedicated, secure and potentially trained engine. And not to spoil the fun, but that’s not exactly what the free online translation tools offer.
We’ve all heard it before: “There’s no such thing as a free lunch.” That rings true for the free online machine translation engines as well. Most of us have heard about employees of the Norwegian oil giant Statoil, who discovered confidential documents that were originally translated using Translate.com, were openly available to anyone who conducted a simple Google search.
Some companies are highly aware of the security issues associated with machine translation such as the Oslo Stock Exchange, which has blocked access to sites such as Google Translate and Translate.com. Other companies are implementing similar practices or applying strict guidelines for employee use of free machine translation to avoid similar security issues.
A breach like the one at Statoil highlights the need for secure translation solutions. But what options do companies have when work is piling up and the translation was needed yesterday?
A line of companies, including us here at LanguageWire, offer inexpensive alternatives to free online machine translation engines. Companies can keep the efficiency and cost-effectiveness of machine translation, while at the same time ensuring their data doesn’t end up in a random Google search.
A highly secure option is an on-site machine translation engine that runs inside the corporate network so there is no transfer of data outside the company walls. Another secure option is cloud-based machine translation that is individual to each company and has secure, encrypted data communication.
As you embark on your machine translation journey, a tip is to look for companies with an ISO27001 certification. This international standard has processes for managing sensitive information and guarantees that if data is lost, there is a robust process in place to resolve the situation.
Free white paper: Optimise Content Workflows with Machine Translation.
Are you struggling to create more content, in more languages, in less time, and at a lower cost? Read this handy guide on how to streamline your content workflows with make use Neural Machine Translation.
Our overall advice is to avoid using free machine translation services when it comes to company documents, text or contracts. The information you are pasting into a system such as Google Translate will stay on a server somewhere in the world, and in that case, you are no longer in control of the information. All companies need to maintain control of their information and the nature of the Internet makes that a difficult task.
Taking measures to keep your business information secure should be a priority and you would hate to see a harmless translation be the cause of a critical data leak. So next time you’re tempted to use a free machine translation engine, consider security and remember: there is no such thing as a free lunch.
Related content
How can we help you?
Your journey to a powerful, seamless language management experience starts here! Tell us about your needs and we will tailor the perfect solution to your enterprise.
