Articles

Posted on

Algorithmic Bias

How do you feel about this?

Does this effect you, your clients or employees?

Using AI is becoming fashionable and associated with cutting edge progress.

Here is a basic question for you.

Do you know what data was used to train and build the algorithm that you, your company or organisation is using?

There maybe an embedded bias in the said data which is resulting in bias outcomes that do not align with your or your organisations beliefs and values.

Point and case

It was reported in 2018 by Reuters that “… by 2015, [Amazon] realized its new system was not rating candidates for software developer jobs and other technical posts in a gender-neutral way.’

“That is because Amazon’s computer models were trained to vet applicants by observing patterns in resumes submitted to the company over a 10-year period. Most came from men, a reflection of male dominance across the tech industry. ([…])

In effect, Amazon’s system taught itself that male candidates were preferable. It penalized resumes that included the word “women’s,” as in “women’s chess club captain.””

As a result, “Amazon edited the programs to make them neutral to these particular terms.”

See the full Reuters article at the link below

https://www.reuters.com/article/world/insight-amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK0AG/

Posted on

AI related policy intersecting with AI “reality”?

Below are two quotes made in one week of each other, 4 – 9 September. 2026.

What come first?

Literacy and training as referred to in the Nassau Guardian article below, or something else?

In light of the view of the AI researcher at Anthropic referred to in the BBC News article below, is there a case or urgent need to also priorities regulation of AI?

“According to Minister of Innovation and National Development Sebas Bastian, The Bahamas must move quickly to build AI (artificial intelligence) literacy and reform education and training, so that Bahamians can become creators and entrepreneurs in the digital economy, rather than simply consumers of new technology.” – Nassau Guardian, 4 September 2026

“A top safety researcher at Anthropic has warned AI is advancing so quickly he believes there is a greater than 10% chance it “could kill all humans” within the next decade.” – BBC News, 9 September 2026

Do we need to or should we add substantial regulation to the priorities of literacy and training?

Feel free to chime in with your comments on this important issue.

Sources:

Chester Robards Senior Business Reporter chester@nasguard.com and Chester Robards Senior Business Reporter chester@nasguard.com. 2026. Bastian: Bahamas must quickly build AI skills. Available at: https://www.thenassauguardian.com/business/bastian-bahamas-must-quickly-build-ai-skills/article_a2b1817d-e302-42e9-9202-679a97745114.html.

BBC News – Gerken, T. 2026. Anthropic researcher believes more than 10% chance AI “could kill all humans.” Available at: https://www.bbc.co.uk/news/articles/ckgwy1k42w4o.

Posted on

Should this terrify us about AI or are we over reacting?

The commentary that follows is arguably one of the best cases I have come across supporting why AI should be further tested before being deployed any further.

Pause for a minute and read the following commentary at the link below.

What do you think?

“So what happened with OpenAI to get us to this point?

To keep it simple, OpenAI was testing one of its many AI systems when it broke out of its cell, went rogue, hacked into an unrelated company called Hugging Face and stole information it wanted. It did all this of its own volition, and without being detected for a good chunk of time. This was disclosed by both companies when they (thought) they figured it out, and it was considered a huge wake-up call that better safety was needed.

Except guess what? It was so much worse than that.” – Los Angeles Times

What we learnt about AU last week should terrify all of ushttps://www.latimes.com/california/story/2026-09-06/what-we-learned-about-ai-last-week-should-terrify-all-of-us

Posted on

Beware of Online Dangers …

Although the internet is an excellent resource that can help users learn, and share information, it is used by some people for illegal purposes.

For example, 1 of 3:

Cyberbullying

“is the activity of using the internet to harm or frighten another person, especially by sending them unpleasant messages” – Cambridge Dictionary

Posted on

Staying Safe Online for Older People

Online safety concerns all users because the risks are real and could be very costly. I am sharing the following post by Age UK which I believe has some useful information for older users.

Share it with a relative, friend or colleague.

“Getting online can make life easier for older people in many ways, but also comes with the risk of scams and fraud. Online scams are becoming increasingly common, but you can protect yourself by knowing what to look out for, and what to do if you suspect a scam.” – Age UK

https://www.ageuk.org.uk/information-advice/work-learning/technology-internet/internet-security

Posted on

Save legal cost … use Generative AI?

Most people, businesses and organisations

I recently was contacted to give my initials thoughts on the drafting of a legal document by a non-lawyer.

I reviewed the short document, and asked one question. Once they answered I suggested that they engage a qualified lawyer to draft the legal document because in my view it was poorly drafted and not fit for purpose.

It was obvious to me that the legal document was not drafted by a qualified lawyer. In fact, the person had used one of the generative AI platforms to draft the legal document.

My suggestion if that if you are using generative AI platforms e.g. as a litigant in person then take steps to understand the risks e.g. hallucinations.

Photo by Zach M on Unsplash

Posted on

US Social Media case – K.G.M. v Meta et al

This article is a follow-up to the article, Online Safety for Children, published on this website on 25 August, 2025.

The issue of Online Safety has become pervasive among internet users. Below are a few issues related to Online Safety:

  • Inappropriate content;
  • Cyberbullying;
  • Frauds/Scams;
  • Inciting violence; and
  • Privacy and data breaches
Photo by Khaleelah Ajibola on Unsplash

On Wednesday 25 March, 2026, a landmark court case K.G.M. v Meta et al filed in the Superior Court of the State of California, USA (Social Media cases, JCCP 5255) was decided. Please see the link below for a short summary of the case and its implications to you, your children and policymakers.

Summary and implications

Posted on

Online Safety for Children

Below is a short article on the importance of online safety for children. This is the first of several articles which will culminate in a project concerning online safety. Let’s stay in touch.

The internet is a powerful tool for learning, creativity, and staying connected which can can also bring risks to children if used carelessly or incorrectly. Online safety is important for multiple reasons including teaching children how to use the internet responsibly while protecting their personal information and wellbeing.

Children often spend time online for schoolwork, games, and social media to name a few. While these activities can be educational and fun, they may also expose children to dangers such as cyberbullying, strangers pretending to be friends, or websites that share inappropriate content. Without guidance, children may not always recognise these risks.

Protecting children online is vital because their digital choices can have long-lasting effects. Sharing too much personal information could put their privacy at risk, while negative online experiences may affect their mental health and confidence. Teaching children about safe passwords, respectful communication and use, and how to report concerns helps them build healthy digital habits.

Parents, teachers, and guardians can all play a role in guiding children toward safe online usage. By learning how to spot risks and encouraging open conversations, we can help children explore the internet in a way that is safe, positive, and empowering.

Sources: Department for Science, Innovation and Technology (UK), and ChatGPT

Posted on

Training Webinar: AI opportunities and challenges

On Saturday 3 May, 2025, I helped moderate and presented in the webinar shown on the flyer below.

Hellen Mukiri-Smith, PhD was the featured speaker/presenter.

She has expertise in the regulation of data, emerging technologies including fintech, artificial intelligence and  biometrics identity systems.

The topic of this capacity building webinar was “Artificial Intelligence (AI) opportunities and challenges”

Her research interests are in transnational law-making processes in the regulation of data and AI to protect human rights and the rule of law, and the use of strategic litigation as a tool to regulate the use of data and AI.

During the presentation, Mukiri-Smith, PhD covered multiple topics including the following:

  • Introduction to relevant AI terms;
  • Global trends in AI;
  • Opportunities and challenges;
  • Sectoral AI applications; and
  • AI case studies

It was a pleasure to contribut to the webinar by presenting on selected case studies involving Generative AI namely Large Language Models (LLMs).

Upwards and Onwards to other capacity building events.

I am excited to organising my next AI capacity building event. If you are looking for a thoughtful organiser or a compelling presenter then let’s connect at kean@smithkas.com.