2 min read

Artificial Intelligence and Generated Content

Featured Image

Robots are creating content. The internet is full of articles about how much content is created by artificial intelligence (AI) algorithms and whether or not they are true, accurate or simply automated techniques to increase the rankings of social media and other web sites. Notwithstanding all this confusion, there are fantastic opportunities to apply AI in business, and increasingly we are seeing many new solutions use AI to automate human-centered processes, including software videos.

You are probably using tools every day that apply AI to generate content, even if you don’t realize it. Gmail with its Smart Compose feature can predict what you are going to write based on analyzing the subject line and your previous content. Microsoft uses AI to make suggestions in Office 365 on how to improve your writing. Using AI to assemble videos from web pages has been gaining traction with tools such as RawShorts, Vidia and others. Even Google recently announced an AI based tool that can take web pages and turn them into videos.

What’s different about Videate is that we are all about making software videos. While taking screenshots and assembling them into a time-phased sequence accompanied by text to speech is possible, the user experience is totally inadequate for marketing, client success, and customer education in SaaS companies. It doesn’t solve the fundamental sustainability problem, i.e. how much work is required when the software changes. As we have discussed previously, this is a huge problem for companies who follow agile engineering with frequent release schedules. Recording the actual software is the only feasible way to make software videos.

Artificial Intelligence, and in particular machine learning (ML), opens up many new possibilities for automating content creation. Prior to ML, solutions used rules-based approaches that were deterministic, i.e. you specify the rules in advance and manually change them over time. With ML, the approach is probabilistic. Over time the “rules” modify themselves based on the content and improve the results.

An example of a rules-based approach is a widely used technique in technical documentation, whereby XML content is transformed into different forms using a language called XSLT. It works beautifully, but as most tech doc people know, it requires specialists to maintain the rules. While very powerful, it can be expensive to support. Fortunately, Videate can work with existing technical documentation solutions as well to generate videos and accelerate the machine learning process.

Video is one of the most expensive forms of content to produce. It only makes sense that we should use all of the tools and techniques available to automate the work, including AI. Videate makes it possible to reduce the time and cost of software video production by several orders of magnitude. And we are just getting started.

Content creation is entering a new phase. Machine Learning makes it possible to automate work that previously required dedicated specialists. How do you know a human wrote this blog? What if it was written by an AI bot?

By Turing Test standards, if you’ve come this far and not wondered that, then the AI would have won. This same goal applies to videos, which continue to take on more of a life-like form. That is the world we are now living in.

2022: The Year of the Content Renaissance

Every company we speak with has a digital transformation initiative underway. There is simply no going back to the strategies of the pre-COVID...

Read More

Top Ten Favorite Customer Comments

As we head into the new year, we collected up our favorite customer and prospect statements, observations, and anecdotes from the past year. Below...

Read More

Build your Business Case with our ROI Calculator

Over the last two years, we’ve learned a lot about automating video production. To help you build your business case, we have developed an ROI...

Read More