Testing has always been a part of the computing game, as reliability and quality demand systems do exactly what users want. From the biggest operating system to the smallest apps, every piece of software needs to be run through a testing process, and this will always be the case. More recently, an increase in the demands we put on computing systems has evolved our processes, however, with some interesting implications for human and AI workers.
Whether looking at security or generative AI, the testing game has changed, setting a new stage for the future of software and hardware development.
Security and the Human Approach
One of the most rapidly growing and necessary forms of testing is found in the realm of security with ethical hackers. As CyberGhost highlights, these are individuals and teams hired to use their skills at breaking into systems to probe a company’s firewalls and defense systems. Ethical hackers don’t try anything malicious once inside but help uncover holes that security developers might have missed. With software becoming so much more complex, these ethical hackers can help ensure all avenues of potential ingress are covered to prevent the arrival of malware, like the growing risks of ransomware.
It’s not just entirely legitimate hackers who have found testing work in software, either. Some formerly on the other side of the law have also been hired by major and internationally respected companies. As noted by PCMag, some individuals made a name for themselves in some of the shadier corners of the tech world and found respected careers that put their skills to good use. Jeff Moss was one such person who went from being a hacker to consulting for the U.S. Homeland Security Advisory Council. Such career paths aren’t uncommon, demonstrating a greater need for experienced testers of software security systems.
A Competition of AI
Artificial intelligence has come a long way in the last couple of years, with systems like ChatGPT and Midjourney growing ever closer to human levels of skills. We’ve reviewed these systems at TechTricksWorld before, where chat programs have become extremely helpful as tools for guiding people in their writing and learning experiences. One of the most important parts of these technologies, which is overlooked, is that humans are not just testing them. AI is also being tested by itself.
Testing from both sides includes many elements, most of which find themselves in the forms of machine learning that we’ve again previously covered at TechTricksWorld. This is where machines can iterate over time, filtering out incorrect or inaccurate results until they finally produce output that achieves their directive.
In AI testing, this is best illustrated in generative image systems. When given prompts, the output for systems like Midjourney creates images but doesn’t deliver its images immediately. Instead, the generative AI is then measured against a discriminate component. The discriminate part of the system tries to spot where the generative has failed, pointing out what it sees as unnatural to be redone. This process then iterates until the generative system finally outputs an image that the discriminative system can no longer recognize as incorrect. These still aren’t flawless, as hair and hands can often show, but they’ve come a long way, and their form of competition/cooperation is a big part of how their results perform so well.
The Human and Machine Team
Both humans and AI have their advantages in the modern world of computer testing, where neither side can manage a perfect result on its own. Instead, the future is one where AI acts as a tool, crunching numbers while humans pick out the elements that machines misunderstand. This can be especially helpful in security, but the combination is useful for practically any form of work which benefits from automation.
Artificial intelligence developing so rapidly means that we’re on the cusp of a software revolution like nothing that has come before. The result for the user will likely be more powerful software, and it’s thanks to an expansion of human and AI testing this software will be made possible. While there’s no risk of the AI taking over, for now, adopting these systems, combined with the best tech minds of the modern era, have put us in a strong position for the future.
Add Comment