In recent years, especially following OpenAI’s launch of ChatGPT, uncertainty about the future of various professions has grown significantly. Almost all of the online platforms are full of discussions like “The Future of [X] Profession.” It is hard to predict how the future is going to be look like in next 5 years for any of the profession as AI is yet to be matured.
In this article, I’ll share my perspective on how the software testing is likely to evolve by 2030, discussing the impact of AI and how the roles which we currently hold going to be impacted.
Before we go into the future, we should know that how do we evolve?
The changes which we anticipate to come in future are generally inspired by today’s problems and learnings. Few years back world had a problem to get the meaningful observations from huge data so we discovered various technologies like machine learning, natural language processing, and predictive analytics. Similarly, a decade back software industry was dealing with a problem of manual testing efforts and we solved it by automation testing. We learnt over few years that the waterfall approach of developing software is not the most productive way of working, so now everyone is trying to be agile, we evolved many tools to support CI/CD practices.
Let me just try to predict the future of software testing based on the problems which face today and learnings which we have gained over the past few years. I won’t be accurate for sure in my prediction but let me share my perspectives on where the industry might be headed.
The Technology
As I mentioned, the future will be the solution of today’s problems and our learnings. At present, I see the biggest challenge of software testing is writing code for test automation.
There are various players in the industry selling their record-replay automation tool loaded with AI capabilities. However, these tools are still half baked, most of them are still expecting human to add code when there is a complex step to automate. By 2030, I am hoping these tools to mature. In upcoming years, test automation is completely going to be turned around, we may see tools which will be fully capable of developing automated tests with minor guidance from human.
In recent years, we’ve learned that non-functional aspects of software, particularly speed and security, play a crucial role in end-user decision-making. Even in developing countries, data security has become a critical concern, with the increasing awareness, users switch the application the moment they find it slow. This is a long-due challenge for automation tools. Looking ahead, I anticipate the rise of more user-friendly and intelligent tools designed to perform functional, load and security testing efficiently from a single code. Perhaps, I am being too optimistic here, but I am personally all in for this solution.
The Role
A decade back, software teams were finding it challenging to manually test a software. So they hired software testers with proficiency in automation. Later we learnt that, manual and automation skills are not enough to ensure quality.
Now the preference has changed, automation expert is expected to be equally good in all facets of software development. Now the ask is to be a quality engineer, who can communicate with business, pair with designers to assess the user experience, conduct bug-bashes, exploratory testing, chaos testing, performance testing and many such activities. Industry had a problem to solve so they kept evolving this role and by the end of last decade, the expectations from this role changed from being an automation expert to be a multi-skilled quality engineer. However, this is not yet reached at maturity, we have just entered in this phase of becoming a multi-skilled qa engineer, upcoming years would be popularising these skills and making it mandatory bar to hire. At the same time, we would expect them to be less contributing in test automation because of matured AI based automation tools.
In last few years with the introduction of platforms like TikTok, Youtube Shorts, Instagram made people impatient. There is drastic change in human behaviour, especially in term of their usage. Most of the e-commerce companies are innovating their application to reduce the think time while order ing a product. That means, we need scaleable, faster but secure software. There will be huge demand in near future for testers who can help organisations to build a faster and secure application. This is another problem to solve in the industry and it will end with QA becoming a multi-skilled qa engineer, equally proficient in non-functional testing.
Rise of AI has also made this space very competitive, I believe the software industry is the only industry which has adapted to AI quickly. Initially there were some reluctance from management about usage of AI in their organisation but now they are happily providing AI capabilities to their engineer with certain guidelines. It means, this space has became much faster in delivering their product. Now writing software is like a race, competitors are coming with new features at much faster pace because of increased productivity by AI. By the end of 2030, we might see software are being developed on the run time based on the human behaviour. AI will become too intelligent to write the application code based on user clicks and potentially software testers would be the ones who will be stopping AI to becoming crazy and building anything which may harm the ecosystem. However, software testers may not be called as “software testers” by that time. That means, we will be seeing QA evolving as an AI expert as well.
Read, Why I prefer GetByRole over ID, CSS as a Locator
Thank you for reading this post, don't forget to subscribe!
