Milton Friedman was a Nobel prize economist. He appeared to refer to the custom of American saloons serving highly salty food to customers who purchased one drink. It increased the consumption of beer by those customers who ate the food. Milton Friedman suggested that the price of the drinks is likely to cover the cost of producing the food – therefore, consuming goods has a cost, some hidden cost. No one has something for nothing.

No Free lunch, no free social media

Social media brings us together – we share our inner thoughts. We can create for free an account on many platforms, generate our own content, and create those online communities. The hidden cost is the loss of privacy, handing in our data to big firms and media for monitoring societal trends and starting advertising accurately on such platforms. Our entries last forever.

Third parties can start merging data about our entertainment, transportation, and other goods consumption. Profiles can be built about individuals, which in the wrong hands, can lead to unwanted consequences and impersonation. Political campaigns can start using our data to accurately target their campaign messages and monitor societal moods. No free social media; the hidden costs can be greater than expected. If you are in doubt, then the novel 1984 illustrates the dystopia we may be walking into.

No free lunch, no free generative AI

Generative AI is based on a century of literature, human bias, and web content. So I have set the AI generator to create an image of a typical data, computing, and math expert. It created a lovely generated image that does not really resonates with inclusion.

The AI generator has brougth pre-conceived ideas from a recent past. It has been observed approximately 30% of STEM workers are women. It just continue spreading potential human bias and misinform a modern inclusive society. This second image may have been AI generated – at least its creator may have added more inclusivity criteria.

The content of this blog is likely to be used by AI generator. My social media content is likely to be used by AI generator. My academic and peer-reviewed publications are likely to be use by AI generator. At least, a AI generator returned the following when I asked Is it a good thing to use AI generators?

Geminy generated this outcome:

AI generators can be a great tool, depending on what you’re using them for and what you expect from them. Here’s a breakdown of some pros and cons to consider:

Pros:

• Boost Creativity: AI generators can help you brainstorm new ideas and get past creative roadblocks by generating unexpected combinations or variations on a theme [1].expand_more
• Save Time: If you need to generate a lot of content, AI generators can do it much faster than you could on your own [2].expand_more This can be useful for things like creating product variations or social media posts.expand_more
• Accessibility: AI art generators can help people who can’t draw or design themselves create visuals for their projects [4].

Cons:

• Limited Control: You don’t have complete control over the output of an AI generator.expand_more You may need to generate multiple images or pieces of content to get what you’re looking for [2].
• Originality: AI generators can’t truly create original ideas, they can only assemble variations on what they’ve been trained on [1]. This means there’s a risk of outputs being derivative or generic.expand_more
• Bias: AI generators can reflect the biases present in the data they’re trained on be sure to use a reputable generator and to carefully evaluate the outputs for any biases.expand_more

Overall, AI generators can be a helpful tool, but it’s important to use them with realistic expectations. They’re not a replacement for human creativity, but they can be a springboard for new ideas or a way to quickly generate content.pen_sparkexpand_more

The AI generator may be useful to reduce search, but it can reduce our creativity too. Old content and bias may be pertained. Hidden costs in editing, managing biais and misinformation should be considered by its users.

No free lunch, no free open- source software

The open-source software has been supporting research communities, open heterogeneous programming languages, and software. OpenStack originated from NASA – it revolutionized the use of cloud infrastructure software. Java, Python, R, and JavaScript are all open-source programming languages used by companies, governments, and universities to build new software products. PyTorch, Keras, Scikit-learn, TidyVerse, Ranger, and DataShield are all open-source libraries that support the use of advanced statistical methodologies and machine learning techniques. Finally, Unix lead the way at the early stages of the movement. All those examples are not exhaustive.

The hidden cost of all this open-source software movement may have reduced the cost of purchasing licenses. The hidden cost lies in finding the appropriate computing skillsets to build and maintain IT resources to bring some digital transformation.

Imagine trying to use specialized libraries without fully grasping their concepts. Developing AI models or statistical predictive models would be quite daunting without a clear understanding of the results of those computations. Consider using various open-source database management systems – like relational, document, and tree-based systems – without properly designing and planning the data from its initial capture to its ultimate use to support decision making. The hidden costs involved in this scenario include the need to involve user researchers, as well as data, technical, and user interaction specialists, to (1) gain a comprehensive understanding of the issue, (2) devise a solution based on data structure, and (3) ensure user usability that meets unique user needs.

No free lunch, no free low-code development

Low-code development platforms offer graphical user interfaces interaction that support business application development through a rapid process and without requiring coding knowledge. While its attractiveness is growing, it may bring some hidden cost too.

Low-code and no-code platforms may be open or closed. Some licensing cost may bring some expenses in relying on software as a service (SaaS) model. The low-code and no-code platform would have made some compromises and simplification of computing concepts. For that reason, some hidden cost may arise to attempt to develop bespoke. Some low-code development platforms would have generalised problems solved by software development; i.e., enterprise resources planning, customer relation management, data analysis, creation of graphical visualisations, assets (data and information) management. This list is not exhaustive.

Some hidden costs arise when an integration with third-parties is needed. The authentication between systems and cyber-security settings require specialism knowledge. The reliance in one vendor brings risks of fluctuating prices or discountinued provision. Finally, unforseen hidden cost to software and data governance needs to be considered.

To conclude

Consuming goods has a cost, some hidden cost. No one has something for nothing. Software is a form of goods that impacts on its users. The free use of social media has let us to trade our privacy, to allow big corporations to profile individuals. Generative AI uses our generated content and bias through literature to generate content. Misinformation may thrive. It may not only costs organisations for misinforming the public, but also society in its ability to differentiate between reliable and unreliable content. Finally, open source or low-code development both rely on specialism knowledge to bring the right software solution that meets users needs. IT, digital transformation, and use of IT system is costly. We have yet to find a free lunch solution.