We all feel it and I felt it today in a massive way (with assistance from IKEA) it is not that I don’t get it, I get it just fine, but I face the aging of my body, it is getting to 70 years old now and varying three piece es of furniture to my trolley implied tossing 6 boxes on that trolley and each and every try one was throwing out my back (and then some). Still, the IKEA staff members were excellent in handing me support and assistance. I reckon that they were thinking “What the hell is he doing carrying 240KG in boxes?” A thought that should have occurred to me quiet plainly. And the next day (whilst I was eagerly destroying my knees) I occurred to me that there is a second setting of bias, bias of my own abilities. So when we (now) look at bias (like Rina Diane Caballar) we see “Humans are innately biased and the AI that we develop can reflect our biases. These systems inadvertently learn biases that might be present in the training data and exhibited in the machine learning (ML) algorithms and deep learning models that underpin AI development. Those learned biases might be perpetuated during the deployment of AI, resulting in skewed outcomes.”Humans are innately biased and the AI that we develop can reflect our biases. These systems inadvertently learn biases that might be present in the training data and exhibited in the machine learning (ML) algorithms and deep learning models that underpin AI development. Those learned biases might be perpetuated during the deployment of AI, resulting in skewed outcomes.” It occurred to me that bis and validating requires a timestamp of this validation, because out via of the outcome is becoming the bias of the equation.
So when we see “Skewed Outcomes: Machine learning models are trained on large human-generated datasets that often reflect real-world prejudices, leading to biased decisions in hiring, healthcare, and policing” which is what IBM is handing us, but the hiring manager is trying to do what a hiring manager is trying to do, rolling with the punches of time, but his ‘assistance wizard’ is not doing this and it becomes a grey/black issue for the machine, whilst the hiring manager is looking for that white resume. So whilst everyone is ‘bottoming his/her resume’ quality is falling in between the cracks of unemployable ad the traps they have.
We then get “Algorithmic Opacity: Deep learning models function as “black boxes,” making it difficult to trace how or why a specific discriminatory choice was made.” Which we get from the Office of the Victorian Information Commissioner. And whilst these models are not to verbal on how decisions are made there is every chance that this will instill a brain-drain. You see, some others can see that some degrees sound impressive but they are useless (in the beginning) so whilst some will drag this out by handing their tech support to philosophy graduates. The hiring manager is confronting a new workforce where sone stay quiet, so will revert to the caveman analogy and we can go on a few turns on that, all whilst the hiring manager requires a programmer who correctly addresses the Machine Learning modules and start preparing for dozens of exceptions and exclusions, whilst the essential inclusions are largely overlooked (actually they are merely overlooked until it is too late to do something about that). And it becomes worse because the Office of the Victorian Information Commissioner will try to explain the wrongful act of Discriminatory choices, making transparency even less likely.
And then IBM gives us a gem (a diamond in the rough if you prefer) “And while organizations are taking advantage of technological advancements such as generative AI, only 24% of gen AI initiatives are secured. This lack of security threatens to expose data and AI models to breaches, the global average cost of which is a whopping USD 4.88 million in 2024.” Does anyone realise that in hiring the lowest assistance can secure over 60% whilst the Gen AI is getting only 24% and that is before some could have thought of “Search for security gaps in AI environments through risk assessment and threat modeling.”, “Assess model vulnerabilities by using adversarial testing.” As well as “Invest in cyber response training to level up awareness, preparedness and security in your organization.” So whilst some might do some of these things, the investment of millions merely hands them an optional rise from 24% to 45% securing investments and whilst corporations are looking towards ‘Microsoft data centres swing to $326m profit amid investment spree’ all whilst they all listen to Satya Nadella talking up a storm, we see “Accounts filed with the Australian Securities and Investments Commission show the company increased revenue 33 per cent to $3 billion, and paid $165 million in income tax for the year” (source: Financial Review) and no one is looking for ACTUAL success stories. I saw one success story, it was BMW with “Cut vehicle production line downtime by 25% while boosting quality inspection accuracy to 92%” but was that actual AI in use, or was that machine language optimizing machines? We see all kinds of numbers thrown at us, but whilst we see ‘dodgy’ accounting in effect like “Microsoft has not laid off 60,000 workers; instead, in late October 2026, the company announced a Voluntary Exit Offer (VEO)—a buyout program—offered to about 60,000 employees (roughly 27% of its global workforce of over 220,000 people” So where are they going to go. And I am still of the mind, when the world wakes up to the flim flam AI artistry, we will see managers falling over themselves to rehire whatever they can and their opposition is doing the same and likely succeed where the VEO offers fails. Because when we see that machines cannot do it all, when the clarity that there needs to be a person who activates the verifying module and when these big wigs come back with the setting our AI does that, and whomever gets a day of looking into that, will see a tsunami of harm coming all their ways. But I have learned that managers who get rid of the people who gets them into hot waters, there is the opportunity for the adversary of that firm, but what am I saying, Microsoft and Amazon doesn’t have any adversaries, do they?
So what does any of this to do with aging? Well, there are a few settings “Data aging is the process of moving old, low-relevance, or historical data away from active, high-performance storage or memory into secondary or historical storage areas” (source: SAP) as well as “It moves inactive or closed records (such as old financial transactions) from hot in-memory storage to cold disk partitions within the same database. This reduces the main memory footprint and speeds up queries for active day-to-day data, while keeping historical data accessible via specific SQL range restrictions (WITH RANGE RESTRICTION).” The question becomes have these training modules kept up with these aging routines? Because that element is often not in exported data, as such the validity of the data as well as its verifying functions are failing. So whilst we are considering “such as old financial transactions”, so how are these records of Gerald Payne, World Plus and a few others are seen by these ‘machines’ and whilst we are aware that there is a link between these settings, the setting of ‘with age restrictions’ might not be used everywhere as such the amount of credible data is failing (harshly) and I am merely stating obvious cases, but these poi scheme settings are still being used and some are ‘confidential’ because they are being investigated and syphoning out these dealers gives them a case for suing for discrimination. Because in the United States (as far as I know) “Under federal law (Title VII of the Civil Rights Act), the Equal Employment Opportunity Commission (EEOC) states that blanket bans on hiring anyone with a criminal record can be discriminatory.” So these people merely need to prod the fake AI system to see where they are ‘disregarded’ and start a case against these discriminatory parties. I reckon that this will be a class action case for 2027. Don’t take my word for it, just watch the news, it will happen. I feel decently certain of that.
So there am I (still aging) taking up my fight with the white Fågelfjället, I will not be defeated by my age or that massive amount of wood (even if it kills me).
