Skip to main content

AI bias

[PLACEHOLDER]
Image

Objectivity can be challenging. Machines are not shielded from this by any means.

As humans we are leveraging technology in a progressive manner, making our lives easier, becoming an extension of ourselves in our day-to-day existence. By trying to simplify things, complexity creeps in (it is ironic, I know!), and as we try to create something for the good, very often it can be used to go against the principle upon why they were created in the first place.

 

Photo by Pavel Danilyuk from Pexel

Therefore, as creators of this technology, usually with the best intentions, it is understandable when model predictions become susceptible to bias.

Those in charge of building the models need to be aware of the common human biases that will find their way into the data used, allowing them to take proactive mitigation steps.

To help remove those biases then we need to have structure. A framework that allows us to build AI systems in an ethical manner that can benefit our communities. AI ethics is that “set of guidelines that advise on the design and outcomes of Artificial Intelligence (AI)”.

AI and bias

Algorithm bias “describes systematic and repeatable errors in a computer system that create unfair outcomes” - Wikipedia.

Putting close attention to the algorithms, and the data, selected for the implementation of the solution “X” allows to identify, prevent and/or mitigate real concerns, such as systematic and unfair discrimination.
AI bias can happen due to cognitive biases, which is an unconscious error in thinking that leads you to misinterpret information from the world around you, affecting the rationality and accuracy of decisions and judgement.

The teams in charge of training the models can be ingesting those biases through the data collected.
Note: The use of incomplete data for data modelling training is another potential way to have AI bias as your outcome.

The use of incomplete data for data modeling training is another potential way to have AI bias as your outcome

ai-responsibility-beolle-blog

AI bias types in ML 

Here are a group of bias types that can be found in data (Go to the Google Developers link in our references section for more details and examples): 

Automation bias Selection bias Group Attribution bias Implicit bias


Reporting bias

When the frequency of events, properties, and/or outcomes captured in a data set does not accurately reflect their real-world frequency.

Automation bias

a tendency to favour results generated by automated systems over those generated by non-automated systems.

Selection bias

Occurs if a data set's examples are chosen in a way that is not reflective of their real-world distribution.

Group Attribution bias

A tendency to generalize what is true of individuals to an entire group to which they belong.

Implicit bias

occurs when assumptions are made based on one's own mental models and personal experiences that do not necessarily apply more generally.

Reference

Trending posts

Steer for a talent transformation strategy (and avoiding AI fatigue)

 There was a debate on whether to feature the term “AI” in the title of this article. Honestly, a key motivation for pursuing the research that led to this post was sparked by the widespread excitement about AI appearing constantly in our LinkedIn feed, to the point of feeling the fatigue, and even a bit disappointed in the algorithm of this, and the others, social media and content curated apps.  We soon discovered that there is an entire concept called "AI fatigue", not exactly how we were feeling it, but more about the mixed emotions people in the workforce have regarding the use of AI tools. Photo by Mart Production via Pexels (background updated with AI and Adobe  tech) From micro blog posts to video podcasts, lately, most of the tech content we encounter revolves around AI. They often sound or read very similar, usually mentioning the same few top providers. The articles (and social posts... at least the popular ones with paid-campaigns behind it) tend to focus less...
Read more

Reimagining Digital Experience Management: How Agentic AI is Transforming Adobe Experience Manager

 Adobe Experience Manager (AEM) has introduced powerful new Agentic AI capabilities designed to continuously improve and adapt digital experiences at the speed of AI. By integrating advanced AI orchestrators through Agent-to-Agent (A2A) and Model Control Protocol (MCP) tools, AEM enables brands to automate complex workflows and enforce compliance seamlessly across enterprise ecosystems. Through a suite of specialized agents, teams can transition from manual, weeks-long processes into fast, AI-assisted workflows powered by simple natural language prompts. Photo by Tunahan KALAYCI via Pexels   Here is a breakdown of the key agents driving AEM’s new Agentic capabilities, their value propositions, their guardrails, and their current availability status. 1. Brand Experience Agent. Overview. The Brand Experience Agent accelerates digital modernization through specialized sub-agents—the Experience Modernization Agent, Experience Production Agent, and Experience Development Agent. Tog...
Read more

Designing Habit Forming Mobile Application

Anonymous
Mobile Applications have become an integral part of our daily lives - we use mobile apps as alarm clocks to wake us up in the morning, to create to do lists when we start our day, to communicate with our colleagues at work via apps like Skype. We even check reviews of restaurants to visit on apps like Yelp and we seek entertainment on apps like Netflix and spotify. So what drives us to use these apps so seamlessly in our daily lives? Why we prefer some apps over others? Is there a science behind designing successful mobile apps like Facebook?  Photo by Peter C from Pexels A study in US revealed that a user between the age of 18 and 44 visits the Facebook app on average 14 times a day [1]. This shows that using the Facebook app is a daily routine for many of its users. This makes Facebook a great example of a habit forming mobile app which is designed with human psychology in mind that encourages habit forming behavior in its users .   I recently attended a seminar ...
Read more

Assembling MLOps practice - part 1

In one of our previous articles it was highlighted how DevOps manages the End-to-End application cycle, leveraging agility and automation. CI/CD pipelines, collaboration and transparency, monitoring and automation are part of the list on how DevOps leverages and facilitates agility. What if then we bring those to support ML? That is how MLOps comes to the table and starts making sense! Lego Alike data assembly - Generated with Gemini A big tech corporation, or a startup, nowadays will see how it is becoming a requirement to incorporate AI and Machine learning (ML) in their operations. ML components are key parts of the ecosystem, supporting the solutions provided to clients. As a result, DevOps and MLOps have become part of the "secret sauce" for success.  What is MLOps Just to bring the definition of what you probably know (or put together based on the above) MLOps focuses on the life-cycle management of machine learning models. It combines machine learning with traditional ...
Read more

This blog uses cookies to improve your browsing experience. Simple analytics might be in place for pageviews purposes. They are harmless and never personally identify you.

Agreed