Category Archives: AI

An Absolutely Fabulous Article by Cory Doctorow on the (Gen) AI Bubble …

and how it’s going to pop like every other tech bubble since the first dot com bust!

What Kind of Bubble is AI?
  by Cory Doctorow

Cory doesn’t say it, but he makes it pretty clear that when the bubble pops, like every tech bubble that has come before, there may not be much less to salvage when it does (especially since no one is thinking about what happens when it does pop).

So I’ll clarify:

A lot of people are going to lose a lot of money

(and while stupid investors hyping this bandwagon heading for a cliff probably deserve to lose every penny, all of the pensioners in the pension funds they scammed don’t; so if you run a pension fund, please pull out of ridiculously overvalued Gen AI NOW!)

A lot of people are going to lose their jobs

(and it’s going to be more devastating to the tech sector than the Silicon Valley Bank failure this year combined with the recession forecast that resulted in over 250K IT jobs being slashed in the USA alone)

A lot of hardware is going to suddenly go idle

and smaller cloud providers are going to go under when the big name cloud providers all of a sudden drop their prices to the floor just to keep the revenue coming in (resulting in the monopolies of Amazon, Google, and Microsoft controlling most of the servers outside of China and Russia)

The problem is, as Cory clearly lays out, when you take one step back and look at the ridiculous hype from a business/revenue lens, all of the big, exciting use cases for AI are either

a) low dollar [and low-stakes and fault-tolerant] (helping us cheat on our [home]work or generating stock-art for bottom feeders [who won’t pay an artist and don’t mind ripping off the IP from thousands of artists]) or

b) high-dollar but high-stakes and fault-intolerant (self driving cars, radiological cancer detection, worker screening and hiring, etc.)

and when you consider the data center costs of these super-sized models (as these data centers consume MORE energy than a small town), low-dollar AI applications won’t pay the bills and high-dollar AI applications cost MORE to deploy than to just do it the traditional way with an educated and capable human!

E.g. self-driving cars don’t work (and “Cruise” needs to employ 1.5 times as many supervisors as a taxi service would employ drivers to keep their cars, which still hit and critically injure people, relatively safe)

E.g. radiological cancer detection requires a human expert to spend the usual amount of time in diagnosis before consulting the AI, and then, if the AI doesn’t agree, spend that much time again

Not that we’re not stopping you from jumping on the (Gen-)AI bandwagon or selling that silicon snake oil that Open AI and Microsoft AI are selling. We’re just not joining you on the (Gen-)AI bandwagon as the steering algorithm is defective and it’s heading straight for a very high cliff at a very high speed …

Merry Christmas!

Good Questions to Ask If Procuring Tools With AI, Especially If You’ve Answered the First Question Wrong!

Continuing on with our statement that sometimes you have to listen to a lawyer, a recent article over on Bloomberg Law noted that Companies Should Ask These Risk Questions When Procuring AI Tools and gave us four questions in particular that were good:

Do I Understand the Data

The article gets it right when it says that AI tools are only as robust as the data they’re trained on, as well as the need to know what data is collected, how, and if all rights are respected when doing so. But what they didn’t get is that the data determines what models and techniques can be used, and what models won’t be that effective or reliable. A vendor sales rep will tell you that whatever technique it’s using is just right for your problem, but the reality is that the sales rep likely doesn’t have anywhere close to the mathematical knowledge to know if its appropriate or not, especially since that sales person may have barely passed remedial junior math (as not all US states require remedial senior math to graduate High School). Furthermore, there’s no guarantee that even the tech teams know if the model is appropriate or not. If the company just hired a bunch of developers with maybe a year of university math, gave them access to a bunch of libraries, and all they did was test out various machine learning models until one appeared to work to a sufficient degree of accuracy on the test suites they compiled, it doesn’t mean they understand the model, why it worked, or even the appropriate characteristics of the data set that allowed the model to work — it just means that they can say for data sets that look like this, it should work. (But what is look like?) You need to understand the data, and find someone who understands the models that it is appropriate for.

Have I considered Regulatory Scrutiny?

Not only do you have to take note that The Department of Justice, Federal Trade Commission, and other regulators are focused on whether technology companies and their tools create anti-competitive environments or put consumers at a disadvantage, but many jurisdictions are considering or implementing laws against the use of black-box technology where the output — which determines whether or not a person can get a loan, be insured, or even apply for a job or government program — and the logic behind the decisions, and the rules that were applied, cannot be explained. You could also be in trouble if the process is fully automated and there isn’t a human in the loop to validate the decision, if the systems uses (third party) data that it has no right to use, or if the output data is not sufficiently protected if it was generated from input data that must be protected and the output can be reverse engineered.

Have I Mitigated Security Risks?

It’s not just traditional cyber attacks on the system, it’s well calculated queries that can slightly perturb the system over time until the outputs after the 10th, 100th, or 1000th slight, imperceptable, perturbation result in an output the system never should have given in the first place, such as approving a ten million dollar loan to a high-risk foreigner who will take the money and run or denying insurance to all people with a genetic defect likely to result in a specific condition down the road that can only be treated by a single drug owned by a single pharmaceutical who will drive people into bankruptcy for a pill that costs $5 to make.

Did I include Best Practices in the Contract?

More specifically, did you include the best practices you want followed in the contract? Don’t leave best practices up to the vendor to define however they want to define them. Make sure you cover all necessary security measures, compliance with all government and regulatory guidelines on AI in the regions you intend to use it (and open standards if there are none, guidelines from the UN, the Responsible AI Institute, or something similar), and so on.

And these are great questions, but the first question you should always ask is:

Do I Really Need AI?

And only when you choose the wrong answer, and say yes, do you need to ask the questions above. The reality is that you don’t ever need AI. AI means that you, or the vendor, were just unwilling to take the time to understand the problem and design an appropriate solution. Remember that when you try to jump on the AI bandwagon heading off the cliff (for the sixth decade in a row).

The first jobs lost to OpenAI were at OpenAI? I LOVE IT!

In honour of the first five jobs that were lost to OpenAI, at Open AI (where it was announced the CEO, president, and 3 senior staff were stepping down and/or let go this week).

To the tune of I Love It by Icona Pop (feat. Charli XCX)!

I got this feeling on the winter day when you were gone
You crashed your car into the bridge
I watched, you let it burn
You threw our shit into a bag and pushed it down the stairs
You crashed your car into the bridge

I don’t care, I love it
I don’t care

I got this feeling on the winter day when you were gone
You crashed your car into the bridge
I watched, you let it burn
You threw our shit into a bag and pushed it down the stairs
You crashed your car into the bridge

I don’t care, I love it
I don’t care

I’m on an Earthern road, you’re in the Milky Way
You want me down on earth, but you’re up in space
You’re so damn hard to find, that AI took over
You said it’d take our jobs, but it f*ck3d you over!

I love it
I love it

9% of Companies Claim To Be Ready to Managed Risks Posed by AI? Bull Crap.

the doctor could not believe the recent headline in Forbes that said Only 9% of surveyed companies are ready to manage risks posed by AI. Because there is no way that 9% of companies are ready to manage the risks posed by AI. There’s no way even 0.9% of companies are ready to manage the risks posed by AI.

Why? Because of the rampant introduction of massive LLMs and DNNs that no one understands, for which I’m sure we’ve yet to seen the last of the abysmal failures, hallucinations, and suicide coaxing. There’s simply no way we can even begin to predict all of the potential errors they are going to make, the risks they are putting us under, the repercussions if those errors are made and risks materialize, and how the risks can be minimized, if not mitigated. No way whatsoever.

Not only is it theoretically impossible to be fully prepared, but when you consider that the average organization is not even equipped to handle regular software failures, how can the average organization expect to handle a software-based AI failure it can’t even predict?

The article, which quoted a recent study by RisKonnect (who are obviously able to detect and protect against most types of risk by using RisKonnect, and maybe that’s why they are so confident they can protect and defend against AI risks, but RisKonnect is for traditional enterprise and third-party risk, not cyber risk, and definitely not AI risk — no one can protect against a risk when they don’t even know what the risk is), did quote some very useful statistics on areas of concern. Specifically, of the companies surveyed

  • 65% are concerned about data and cyber,
  • 60% are worried about employees making decisions on erroneous information,
  • 55% are worried about employee misuse and ethical risk,
  • 34% are worried about copyright and intellectual property, and
  • 17% are worried about discrimination risk.

The risks are the right risks, and the order of priority is about the right order, but the percentage of companies concerned is much too low.

1. 100% of companies should be concerned about data and cyber. Not only are we in the age of state-sponsored hacking, which makes any company with useful confidential designs and information a target, but with almost all significant commerce being conducted online, all companies are a target for financial fraud.

2. 100% of companies that need to make decisions based on data analysis should be concerned about erroneous information, as all companies have bad data, and the bigger the company, the worse the data.

But none of these match the risks of AI. As per the quote in the article from Caitlin Begg, an over-reliance on AI can risk robotic, insensitive, spammy, or off-topic messaging, and that’s just the beginning. As noted, most companies haven’t simulated their worst case scenario, and since one can’t even predict what that is with AI, they aren’t even close to ready. It’s not just another article in the organization’s tech stack, even though the article seemed to indicate it is. One can prioritize transparency, accountability, threat and vulnerability monitoring, and risk mitigation, but when most AI applications can’t explain their actions, aren’t accountable humans, have no realistic threat and risk assessments, and there is no way to mitigate risk except not to use the technology in the first place for any decision that should be made by a HUMAN, it’s just not enough.

The precautionary steps are not to identify where AI can be most effective and incorporate it, the steps should be to

  1. identify where partners and third parties are using AI and putting your organization at risk
  2. identify where employees might be using unapproved web-based AI applications and put a stop to it
  3. identify where your SaaS providers are not only using, but introducing, AI into their applications after purchase and delivery and ensure that any utilization is bounded, tested, and properly constrained to prevent risk

Then, instead of unbounded AI, identify appropriate automation technologies that can be properly configured, integrated, and managed as part of an enterprise stack. And reap the rewards while your competitors deal with risks.

Do you want to get analytics and AI right? Don’t hire a F6ckW@d from a Big X!

Note the Sourcing Innovation Editorial Disclaimers and note this is a very opinionated rant!  Your mileage will vary!  (And not about any firm in particular.)

Now, I’m going to upset a lot of people with this, but I don’t care because the linked article below is literally the best article I ever read on why you should NOT hire F6ckW@ds from Big X (or any other) Consulting Firms who claim to be analytics and AI experts when they don’t actually know

  • the difference between a mathematical formula to calculate the center of gravity of a falling object and to calculate the median spend in a category
  • proper software architecture
  • proper compute resource allocation
  • your business
  • the difference between real ML technology, RPA and a few formulas, and the current Gen-“AI” where the “AI” stands for artificial idiocy

because

  • you’ll spend 3 years and millions of dollars to implement something that should take 3 to 6 months
  • you’ll spend hundreds of thousands on big vendor software licenses you don’t need
  • you’ll spend hundreds of thousands on compute power you don’t need

After all, these guys and gals get paid by the hour and the commission on the resell license is a percentage of the total price they convince you to pay for it. So, the longer the project takes and the more licenses and compute power they sell …

Read the linked article. Twice. And then tape it up to your fridge. The situation described in the article is NOT the exception. As a former CTO and 25 year consultant/analyst, I know this is the norm!


I Accidentally Saved Half A Million Dollars
 

Now, if you’re wondering how to tell who is a F6ckW@d and who’s not when it comes to analytics and AI at the Big X, I’m sorry to say that it’s not so easy (especially when it only takes a few bad apples to spoil the bunch, and while the good firms will do mandatory pruning of the consulting tree annually to weed those bad apples out, you don’t want to be the unlucky client who gets one on your project) .

It used to be if they were there for more than a year or two, their was a possibility that they were, or at least not as good as they claimed to be,  that especially if they were junior, right out off school, no real experience. This was because, first of all, tech talent wants to go either to the big glorious tech firms (Alphabet, Meta, etc.) or the wild-west startup frontier, and big consultancies were the backup until they got enough talent to move on.

Thus, the real talent in tech and analytics, who didn’t get promoted quickly in the Big X, usually didn’t stay long before they moved on to specialist firms where they felt they were more respected, higher up, could control the projects, and, more importantly, being higher up, were higher paid.

(Tech/Analytics people take pride in their work [and not their title], and seek the job that gives them the most pride.  Also, even though good tech/analytics people won’t contradict managers because they want to be important, and will only contradict managers because they want the job done right, the reality is that junior people or new hires in big firms often have the impression that this is discouraged in a larger firm [even if it’s not] where you are supposed to learn from and follow your manager’s lead because you don’t see the big picture and may not speak up on the way a project is being approached when they are unsure.  They might be wrong, and should stay quiet, but they don’t learn if they don’t ask.)

However, now that all the big firms are acquiring mid-market experts, with some of the Big X acquiring 3 or 4 specialist plays in analytics and AI over the past couple of years, it’s much harder to differentiate if you are getting the best talent or not.  You have to vet every candidate.  Not the Big X.  YOU!

And you need to remember that some of this AI and analytics stuff is literally so complicated that you need degrees in mathematics and computer science and sometimes a decade of experience to get it right! (It took the doctor two advanced degrees and building advanced analytics and optimization systems for multiple leading companies in the 2000s before he really understood the art of the possible and, more importantly, what was relevant for an industry and what was not.)

In other words, it’s okay if you don’t really get it as a manager. Just find those one or two people who do who you can trust, pay them well, and let them do what they need to make your department look good (be it hire internally, choose a consulting firm you never heard of, hire former colleagues on short-term contracts, use their contacts to get the right person at the Big X, etc.).

They’ll get the job done right and be quite happy to let you take all the credit IF you give them regular raises and a bonus any time they do particularly well. Just put your ego aside and let the people who get it make the tech/analytics decisions, and everyone will win!

But, whatever you do, don’t throw a poorly formed project description over the wall in advanced analytics and AI to a Big X (or any other vendor) and expect good results.

If you don’t know what you need, why, and how you expect to get it, instead focus on what you understand and Use the Big X firm for all of the things you know it is good at, understands implicitly, and has the history and experience to figure out simply based on the type of company you are.   Used appropriately, like any service provider, a Big X can deliver amazing value.   See the linked article on when you should use Big X in our opinion.