I Use Machine Translation. Is My Data Secure?
Mankind is producing more data than ever before. To paint a picture of what is happening here, in 2017 we produced about the same amount of data as in the past 5,000 years. And the size of data we are generating will continue to multiply.
This comparison is astonishing. Especially the fact that from my business’s perspective, we are also talking about the ever-increasing amount of data that needs translation. In fast-developing, global industries it is important to be able to make decisions on the fly and yet with as much information as possible. The main verticals include health care, automotive, biotech, and cybersecurity. The winners are those who have information available as soon as possible and in the right language.
Enter Machine Translation
That is why machine translation, is making such great advances in our industry. If it is deployed and used well, machine translation can save a lot of time and cost to any organization. And it will continue to do so in the future. Neural machine translation, the type of translation we use now, continues to learn, and provides greater benefits. These include shortened cycle-times, lower cost, and greater accuracy over time.
Customers usually have a good understanding of the fact that professional machine translation does not utilize free translation engines such as Google Translate. Instead, it relies on paid and trained machine translation engines. There are elements of artificial intelligence that even help select the right machine translation engine for the domain of the document after “reading” the content.
At first, customers ask questions like “How good is machine translation these days” or “How does machine translation work”. But the very next question that comes to mind is whether the data machine translation processes are safe.
And it is a question we should be asking. With the volume of mission-critical data being at an all-time high, the way an organization operates, including the way it processes translation, presents a challenge.
The Cost of a Data Security Breach
There are several studies out there that show the financial impact of data security breaches for organizations. According to an IBM Cost of a Data Breach Study, for example, the average cost of a data breach in the U.S. is $3.9 million. More than 60% of respondents say they expect at least one attack within the next year. The most affected industry is health care.
This data shows that any global enterprise needs to have a machine translation solution designed specifically for corporations. Moreover, such a solution needs to put security front and center.
Such solutions also need to be available on-demand and must be easily accessible. The largest risk still lies in employees defaulting to using Google Translate and other free solutions because they are the easiest ones to get to when there is an immediate need.
Any such use of a public portal carries significant risks. The “fine print” in the user agreements of such tools always includes language that specifies the provider is entitled to a number of benefits. These include storing, modifying, reproducing, distributing, and creating derivative works using the content you put in. You see, the content helps the machine improve through its learning algorithms. But it is also very vulnerable to hackers.
Security of Machine Translation
So, what security criteria should we consider when thinking about a machine translation solution for a corporation? Here are just a few examples:
- Security, including encryption of the data
- Deployment solutions that work for you – in-house, cloud, etc.
- Rest API for connectivity of any proprietary systems
- The ability to adapt each MT solution to you as a specific client
- Integration with the primary infrastructure you use
- High-quality, human services such as terminology-building or post-editing integrated within the system
- Automated workflows that aid in minimizing “out of the system” communication and data transfer
Machine translation solutions that cover all these points do, of course, exist. And they will only continue to develop and grow as the digital environment in which we live and work is unlikely to change much in terms of the challenges it represents.
As far as translation, we just need to shift our mindset a little and consider security right along with the other top expectations we typically have, such as speed, accuracy, and cost.
Would you like to know more about this subject or available solutions? Email us at email@example.com – we would love to speak with you.