Maximizing AI Impact in Learning and Development and Enterprises

By

-

June 22, 2024

Maximizing AI Impact in Learning and Development and Enterprises

Are you aware?

There is so much out there about AI that it would require someone to read a lot of publications a night just to stay current. They would call him a fool—and yes, I am such a fool.

The good news is that you don’t have to—because I will continue to espouse it for you—and how it can apply to L&D and Training at the minimum and then those pesky enterprises that dove headfirst into AI without realizing all those interesting nuances.

For some reason, there are people who think that I can only talk about the negatives and cons of Gen AI and lack deep understanding and insight into various LLMs, RAG benefits (including what a RAG is), and so forth. For those fine folks, gear up; we are taking off (not on a Boeing, though—Ha, I made a funny).

On a side note, there are vendors in our space that are using RAG, which, depending on what you are planning to do within the system with Gen AI, would make sense with the LLM or LLMs they have. Ditto for companies who might have a RAG, too. You do not need a RAG, BTW.

Studies – It is the land of confusion

I am amazed, okay not really, on the number of “studies” that have come out around companies’ usage of AI in their budget, impact of financial, usage and so forth. Some contradict each other.

The studies tend to reach out to large numbers of companies, as a whole. The way to look at this, is to recognize that relying solely on one study isn’t the best way to ascertain what is going on with AI.

Study A – Lucidworks – Based on responses from over 2,500 businesses located all over the world

Implementation of AI: 25% have not deployed. They are in a holding pattern, stuck in beta—”like a plane circling an airport for 25 days or more.”
Concerns cited by companies around AI in their businesses – hallucinations, data security, and the high cost to run the platforms (foundation is the general term instead of saying LLM, which uh is the foundation)
63% are planning to increase spending on AI in 2024, which is down from 93% who stated they planned to increase spending for 2024
36% plan to keep spending “flat” on AI
And the kicker – 42% have yet to see any financial returns on AI adoption of products

Study B – Slack Workforce Lab

(Polled over 10,000 global desk workers – full-time (30 hrs or more per week – mostly executive, senior, middle and junior management roles), who work with “data, analyze information or this creatively per Workforce Lab’s methodology.”

In other words, global desk workers are a bit misleading – because it’s not Barnaby in the cube next you, unless he was a manager or exec.)

23% usage increase with AI in the workplace (since Jan 24. The study does not provide a to X date); however, the increase of AI in the workplace since Sept 23 is 60% (again, no info on what date this is up to?).
81% say AI has boosted their productivity and quality of work
93% say that work-related issues and trustworthiness – have concerns

The study presented additional information on employee engagement. Again, the results are misleading because it is not employees per say, it is specifically folks in management. Anyway, here it is:

18% increase in work-life balance
23% able to manage stress better
24% increase in happiness in the workplace (the study uses the term – overall satisfaction)
29% who state they feel very passionate about their work
73% “of desk workers – again, that isn’t workers, that’s management” believe the hype of AI is warranted
55% of workers in the age bracket (18 to 29) are highly enthusiastic about AI and the automation to handle parts of their work; only 33% in the age bracket over 60 feel the same way

What the study doesn’t provide is how many of these folks are/were aware of hallucinations (and thus checking before just accepting the info), AI bias just to name two big ones.

Another odd thing about the survey is the age bracket 18 to 29, how many of those folks are senior execs or at a minimum of their management level?

Speaking of execs and AI

I tend to find a lot of execs who have no idea about hallucinations – i.e. fake or false information – especially when it is only their content going into the LLM – hence private.

The assumption being if it is only ours, no issues with fake or false information. That is 100% wrong.

Worse, if you decide to remove your content and untrain the LLM, then put new stuff or updates or whatever back in, you will still have remnants of the previous data in the LLM. – Surprise!

Study C – Accenture

Companies that “apply AI – Generative AI) expect to see higher revenue of 25% after five years.”

For me, I would expect a higher revenue than 25% after five years, simply because AI is at a very tiny stage here, and it will only grow – equally though – not all LLMs are the same, and those darn token fees will hit your bottom line – depending on the number of folk using it – let’s say it is the text version only (like GPT-4o – aka GPT4 Omni, which is free for X number of instances, but the paid version – which would be for Enterprise, increases the number.

When GPT-5 rumbles out, again, for Enterprise in this case, you still have token fees).

On the flip side, if you are using an LLM that is multimodal – which 5, and there are others as well – not just OpenAI here – you can do far more than just enter text and get a retort in text.

As an Emeritus Professor of Economics from Oxford told me – “If you are a risk adverse company, I wouldn’t recommend using AI.”

As folks are aware I often talk about the impact of job losses with AI (as it evolves). While there are plenty of people out there who disagree, saying it either won’t be that bad, or it won’t be the case because there will be more jobs.

If you are in the financial services industry

Citigroup says 67% of banking jobs have “high potential” to change or be outsourced by AI. On the flip side though, AI could add 170 billion dollars to the profit “pool” in the banking sector by 2028.

The article adds that they expect no reduction in headcount – once AI related management hiring is identified and resolved.

One item I found of interest from the Citigroup report is their statement that the banking industry will be slow to adopt due to various factors including regulation and the lack of global rules (which is very big).

Study D- Booth School of Business (University of Chicago)

GPT-4 does a better job of analyzing financial data including statements and predications based on those statements
Using “Chain of thought” which directed GPT-4 to identify trends in the financial statements and calculate different financial ratios. They also found that by using a LLM (Large Language Model), GPT-4 achieved an accuracy of 60% compared to humans that is in the low 50% range (relating to predictive accuracy)
Relating to financial acumen GPT-4 produced more profitable trading with higher share ratios (and alpha) beat the stock market.

Did you know?

Sam Altman, CEO of OpenAI, in a discussion around GPT-5, stated that GPT-4 is the “dumbest model any of you will ever have to use again.” Yes, he was comparing it to GPT-5, which is expected to be far superior, nevertheless his reference could have been something like, “GPT-4 was a solid model, but with 5 coming out you can expect superior results, and blah blah.” He didn’t.

The issues with 4 (which Enterprises were using, and maybe still are, without using Omni) included hallucinations and performance – which was awful.

Numerous people found that in the earlier part of the day, the performance and responses were fast; by the end of the day slow or even the “I can’t respond”.

How bad was it? A study found that the model went from 97.6 accuracy on one test in March 2023 to just 2.4 percent by June.

Another issue was the use of DALL-3 whose output went from the ability to save it as a .PNG to now a WebP format – i.e. a web page. The latter requires you to have software that allows you to change the format into PNG.

Open AI was well aware of this early on – after it started to happen – on their own Developer’s forum, people were complaining about it (and made me feel better, because I thought it was something wrong with my computer). As a fun bonus, Open AI still hasn’t fixed the issue.

The Foundational Model

There are two types of foundational models – the one most folks are aware of is LLM – which means large language model. A LLM requires a lot of computing power to such an extent that it’s carbon footprint is very close to the output of a coal plant.

On top of that, to cool down these computers you need water (a vital commodity in 2024 and beyond).

SLM – is a Small Language Model. An SLM will have a smaller set of parameters compared to a LLM, but it can run on a mobile device, even a laptop.

The latter one intrigues me the most when a vendor has a mobile app whereas folks can access content within the app.

Apple has a SLM in their iPhones, even the ability to use the SLM offline. The idea that a vendor in our industry, including learning technology could use a SLM then, seems possible. Phi3 from Microsoft is just one SLM.

The plus of an SLM is the privacy concerns are seriously reduced, they (SLM) can match an LLM in select functions. Even OpenAI recognizes the opportunity for an SLM themselves.

Be aware though that even with an SLM you will still have hallucinations.

Benchmarks

Folks always, okay a lot of people are wondering “how do they know which LLM is better as this, or better performs at that.” Hello, benchmarks.

Here is a short list of the top ones. Always remember though that they base their output on specific items – which they note – like solving a math problem, for example.

Equally remember just for an FYI – that every LLM, even SLM out there, will do things better or worse compared to another. For example, one LLM does a poor job with web links. Another has an issue with summaries.

Benchmarks to Check Out – Click on “Leaderboard” – Some require you to download files, and are complicated. AA, LiveBench, LMSys are easy to use and read.

LiveBench
AGIEval
Hugging Face – Offers various Benchmarks that sit on Hugging Face
LMSys which uses Elo Ratings – Includes GPT 4o (latest LLM from OpenAI)
Artificial Analysis – I’m a fan of this one, because it also shows the latest from Open AI and Anthropic (3.5 Sonnet, matched up to others with performance as one item, fees as another. Best Benchmark out there, IMO.

Autonomous Agents

This is the GOLD. I wrote a year ago, that autonomous agents were the way to go, and the game changer. Well, read any articles these days around auto agents and you will notice they all point to it being a game changer and its powerful impact.

To reduce the number of times I say Autonomous agents, I will refer to them as either “auto agent” or “agent”.

An agent is far more than a copilot. They can make decisions, are self-directed (i.e. no human input) using chain of thought.

Based on an objective – whatever that objective happens to be (limited constraints – for example, pre-defined APIs – if you were using APIs – Auto agents could make those calls for you and collect the data.

On the other hand, if you just provided an objective, the agent can take over and do a lot. From creating the tasks, to remembering the task, change the task, loop themselves, build-upon said tasks and complete (solve) that objective.

We are talking about self-directed behavior with intelligence. Think this way, they think for themselves (some would say a scary thought) – independence without any ongoing input from you.

From a learning and training angle the capabilities of an agent are endless depending on the objective(s) set forth.

On a learning system, regardless of if it is being used for onboarding, or other employees in general and whatever tasks or things they have to do; to customer training/partner training/B2B to associations, heck even students in higher ed, and K-12.

Is there any vendor working on having agents in their system? Yes. One vendor (name retracted) is working on an agent as we speak.

They expect a launch later this summer or early fall. As an added bonus, when it launches, I will be writing a product review on the agent(s) capabilities along with all the other items they have today in their LMS, and what is coming soon.

What is cool about this vendor is today, they are using AI for skills mapping to content. I always have believed this should be a no-brainer for any vendor, and yet automatically skill mapping with the use of AI, is not an everyday capability among vendors.

Bottom Line

Whether you go with Perplexity (I no longer recommend it), or Claude from Anthropic – Claude 3.5 Sonnet beats GPT-4 Omni (according to Anthropic) or Gemini from Google (underwhelmed, work in progress) or Mistral Codestral (receiving a lot of capital- investment – the highest seed round in Europe – 2023) or Llama 3 from Meta or Falcon 2 11B or AutoCoder Assistant which beat GPT-4 Omni (the latter three are 100% open source, no token fees) or GPT 4 Omni or Nemotron-4 340B from Nvidia or SLMs such as Apple Intelligence SLM or Microsoft’s Phi-3 series or Google Gemma or Gemini Nano (underwhelming) and the list goes on.

Perhaps you want to test a few – the majority have a paid subscription option = usually $20 (USD). I like You.com – which has the latest models, albeit they add their own model with it. Nevertheless, I found the output to be strong – performance wise, and a plus to check out the newest as well as older.

Maybe though you want a Model as a Service (MaaS) instead. These are cloud-based platforms that contain a lot of LLMs that you can pick and choose from. They offer a lot of pluses for folks compared to say just going with an LLM.

There are quite a few out there – I’m a fan of Bedrock from AWS, others like Google Modal Garden. Be aware that with Model as a Service, each of them, regardless of the provider starts with machine learning. For example, Google – its Vertex AI (machine learning) which you need first.

If you are a vendor, I strongly recommend that you have at least two LLMs, rather than one.

Multiple LLMs can leverage their pluses to your system, reduce hallucinations (but they are not gone – i.e. no 100% removal here) and offer some serious advantages. There are vendors out there who have multiple LLMs. Heck one vendor uses three.

The point of this post is to give a wide perspective on AI, from various LLMs to SLMs to MaaS to latest studies, autonomous agents, benchmarks and even the top three learning system vendors using AI.

The list?

CYPHER Learning
Cornerstone Learning Management – Part of Galaxy (And all the items/mods that it comes with at no additional charge – to see the other items, click on Platform, then look under Cornerstone Learn)
Thought Industries (A system for customer training/learning/education – whatever term you prefer)

As for me? Well, FindAnLMS, my platform will launch it’s AI version in 2025.

See you then.

Wait, see me on my site – yep, it has AI too – anyone up for synthetic audio with zero latency?

Get ready for an exciting time. AI here we come.

Learning and Training Leading the Pack.

HR?

Uh, let me get back to you.

With an autonomous agent.

E-Learning 24/7

Other AI Insight

Did you know?

Gen AI and your Workforce

Craig’s Gen AI Group on LinkedIn – Join and get the latest from me!

AI in the Workplace – How Accurate are the Results?

Enhancing Online Learning Adoption Through Creative Marketing Tactics

By

onlinefreecourse

-

June 22, 2024

0

Enhancing Online Learning Adoption Through Creative Marketing Tactics

I’ll be honest. Writing with a bruised bone in your wrist on your writing hand is about as fun as getting pennies in your Halloween bag as a kid. It’s totally not. But today, it isn’t about me – thankfully, rather, it is what I hear so many times, and honestly, for many of you, well, it’s about you and what you are not doing but should.

Seven percent, five percent, less than 10 percent, 25 percent. These numbers are often cited in terms of usage by learners (in this case, employees and, yes, customers in some areas). “I give them access to content, and nobody is using it.” “They get a variety of courses to pick from, and maybe go once and then no more.” “They do not complete the courses, let alone the assignments.” “They really want to be held hand to take the courses.” “It is the system’s fault because nobody is taking the courses that are available.”

100 percent. Over 90 percent. 95% percent. These numbers are cited when it comes to completion rates. The folks that love to quote this, heck, many times vendors themselves, are ignoring the key PIECE here. All of those numbers are tied to one little, tiny thing – IT’S ASSIGNED LEARNING. Think about that. You assign an employee to take a set of courses to learn a new skill or skills. You assign a learner to take some content, i.e., courses for a job role or for, say, proactive compliance or regulation. You assign them assignments. Assign them to complete in a series of cohort courses. You assign them to complete a course using synchronous-based learning. Why wouldn’t you then expect anything less than 100 percent completion or even 90 percent?

Those completion numbers for assigned learning, really mean nothing in terms of learning intelligence. Oh sure, you can spin and quote it to the higher ups as prove this is working. And yes, you can espouse it to prospective clients as validation. Sure, you can leverage it for ROI. “See, this works.” “Learners prefer assigned learning.” “Employees learn only with assigned learning.”

What a load of garbage. If you truly believe that assigned learning is the only way to go, and by doing so, you can validate a high completion rate as proof that employees or customers want it this way, please feel free to resign right now your role as head of L&D or Training. I say this in earnest because nobody – not employees, students, members, or customers wants to be assigned courses. Do you prefer to be assigned learning, when you attending your college or university? Do you prefer being told by a significant other that each week you have to take out the trash, even if it is snowing outside and you have a cold? Do you enjoy the assigned requirements of going to a set of seminars because someone said you have to, even if you have zero interest in doing so?

If you answered yes to any of these, then I bet you give kids pennies at Halloween. The only assigned that should be (albeit there are other ways to do this), compliance and regulation, whereas everyone knows people wait to the last minute to complete it, and nobody is going to go back and refresh themselves by taking the courses.

There is zero proof out there that people will retain over 90 percent of courses in six months by assigning and never returning anymore. As a former director of training, I had assigned learning only for a certification program, and even then, they could pick the courses in any way, not you have to do this first, then that. Even then, it was only assigned and required to be completed months away – and yes, it tied directly into the certification program, which was meant to be challenging. Otherwise, it was pick and choose what you want. Oh, yes, when I provided training to employees online, with compliance, they had to complete it by a certain time frame – would that be assigned? More of we have to protect ourselves in case someone sues because they claimed they never received training on Y, when in fact they did and we have proof.

I’ve written numerous times about assigned learning and faux completion rates. What this post is really about is something a lot of folks in L&D and Training and what other departments oversee online learning often ignore – selling the system to their employees or customers in terms of wanting to get into the system, take the course or courses, over and over again. Regardless of completion – rather because they want to, they like it, and the best way to retain is to align to interest and repetitive experiences again tied to usage, even if they come in 20 times and focus on one chapter. That tells me they have an interest there and want to learn and attain.

How do you achieve that?

When I ran e-learning programs, my usage was always over 90 percent with employees, and that was without required completion. At the customer usage, over 75%. Association members – certification program 100% (of the non-required), association high – (it was a trade association).

The Steps to get there

It’s not going to be easy. Nor should it be. It involves a strategy – think game plan ahead of time – and needs to be ongoing – strategy – pivot. You will always pivot, because even with a strategy – some thing will get the results you want, other times, not so much. That is why those learning intelligence metrics I often cite as crucial, are well, crucial. Tie that into your brain power and watch what happens! After all you were hired because you are an expert in L&D or Training or whatever department e-learning is placed under. An expert. Never forget that. Some folks get into the imposter syndrome when it comes to online learning and their exec role. That is why I always tell people – you are the expert. Not your boss. Not the CEO, nor MD. You. If they were the expert in learning and training, why would they hire you?

Begin

Okay you have bought the system, it is going to take a few weeks or months to get started. There are other items you need to do from adding people (once done), developing or deciding on what content you will add.

Here we are going to multitask. Here is where your strategy begins for those who focus first on the above, rather than thinking about it beforehand. You can choose either route. Sometimes I did the latter, but had the ideas in my brain, and noted on paper; other times, I did it earlier, because I already had the plan of success prior to, and thus, I went about a dual mode – find the system, jot down some early strategy.

What you are going to do is create a marketing plan. Woo you say, a marketing plan? I’m not a marketer. I have no idea on what I am going to do, let alone, how to implement it? Plus, I have everything else I need to do. The marketing plan isn’t tied around creating marketing strategy that someone running marketing at a company – hired to do so, because they are an expert in it. Rather, it is saying, “Okay, what are some things I can do to gain interest and build mass.”

That’s the key – build mass. On average it takes three years to build high usage when it comes to say association members. And it is why you always want a three year contract, and no more. The three means by three years. Some folks might achieve it before then. On the customer side if you are selling courses/content you still need a marketing strategy – this is after all, to generate money. The more you make, the better the budget next year by the company, association, etc. And yes, you will create a strategy for association members.

On the association member side to build early excitement, I created a flyer that went out to every member via snail mail, with the theme of the seven wonders of the world, and the outside looking like something you would see in an adventure movie. Open it up – seven wonders of the world (yes, visually) with each one representing a variety of stuff we were going to have in the LMS. Courses. Video content. Certification and so forth.

Why did I do this?

To get people interested, even excited about seeing this system, to build that early mass. You want people intrigued. Intrigue means interest which means access. Then all you need to do – is have the content (courses, etc.) that is in there, that gets people wanting to take whatever – not just regulation/compliance. And you want them to come back – regardless of their job role, or skill interest or education level. Customer. Employee. Association member.

Your goal is to get that interest going early on. Think of what you have seen or heard of that gains your interest before it is launched? That my dear friends is due to marketing. Build up the excitement, intrigue, and then launch.

Propaganda Techniques

At the same time you are doing the above, you are going to learn about effective propaganda approaches. Here is the kicker – you see and hear them all the time, and just don’t realize it. Why do advertisers do it? Because it works. Why does your grocery store or car dealership or big box online and so forth do it? Because it works.

You remember my little flyer above? I used a propganda technique. With my ongoing marketing strategy I tapped into various propaganda techniques. Yes, once you launch, you need to continue with that strategy. Without over blasting folks emails or fun marketing pieces or whatever. That turns people off – and that isn’t what you want.

Here is a list of propaganda techniques. And below are the ones I found worked the best and how I utilized them – because one of them is by far in my experience a big one – you must do. No exceptions. It works if you do it correctly.

My Picks – Recommend On going basis – One though I use after the system is launched and content (i.e. courses today) is available. Not assigned.

Bandwagon – Oh this one is wonderful. Everybody is doing it so should you. In sports you might hear the term. People are jumping on the bandwagon. Even if you do not have mass, let alone constant usage – who really knows that? Get it? Focus on a few courses, not the entire catalog. Focus on the system itself. “Did you know we offer Chicken University, the best place to find a variety of courses, including how to use AI in your job.” Never BTW mention the words LMS, LXP or whatever type your learning system is. It confuses people. Stay on target – the name of what you call it. “Widget World, FunLand University, etc.”

Bandwagon gives you a lot of options at various times. But, it should be limited. Over usage is bad, and will turn off eyeballs.

Plain Folks – Just like you and me. It is not the samething as Bandwagon. I love this one, because if done correctly, wow, it works and works well. Besides testimonial – which you often see on learning system, even non systems web sites; plain folks delivers – and here is how you can do it – well, how I did it.

Plain folks is just like you and me. Let’s say you have some courses for your executive assistants or those folks who have a similar role. Do you want their manager to push out the value of using the system and some of those courses? Heck no. The goal is “just like you and me.” Plant workers – who are going to select? A similar role in the plant. Or the plant workers themselves – not their manager or someone higher up.” Plain folks ties directly into stakeholders – and that is the goal, I repeat the goal here.

Why?

Who are you more likely to trust? The head of a department, a senior role or someone like you? I’m an exectutive assistant – you have some courses around the latest versions of Office365. You want others in that job role to utilize and see the benefit of doing so. Thus, you are going to find someone in that role who is willing to become your cheerleader. Your stakeholder. They love those courses. They go in quite a bit in X or Y. They see the benefits. They recommend others in their job role to go in and or take the courses.

You are in sales. There is a course on how to create battlecards. You want others to take it, or go in, over and over again – always remember the other goal – usage. Why would you care if they go into the course in June and select chaper Y, 15 times more than others who only go to chapter three? Because when you are creating content or buying 3rd party you can zero in on that topic they are zeroing in on. Who is going to be your cheerleader? Not the sales manager – unless that role is the folks you want to go into that course, over and over again. If the target are regional sales people – then find the stakeholder there. If it is new sales people and the course is cold calling – then find someone similar to that role – a new employee that has gone into the course – wherever it is, and is willing to become your cheerleader.

Get it? This is by far my favorite. Cheerleaders = stakeholders.

Testimonial

I remember the COO of a company I worked at, who went into the system to check out some of the courses. They reached out to me one day and told me how surprised they were because they really liked course A, B, etc. They went back over and over again. That is my testimonial. They would push out a quote for me to use internally. Why they loved the system. Why they loved taking a variety of courses. There goal wasnt to say, “you have to go into this.” Nor “I loved this because it is ideal for your job role or to learn a skill.” Nor “they recommend it.” – All of those come of as disingenous. Rather, they just want their employees to know how much they enjoyed it – how they weren’t sure if it would benefit or they would use it. And to their suprise they are using it, and picking the coruses of interest to them.”

Which comes off as more genuine? Recommend or Surprise is they liked it, and continue to use it? I say the latter because how often do you see recommended, then you go in, and are disappointed. Plus, you might say to yourself, “Well of course they like it. They are the COO.”

The rest of the techniques I ignore. I mean if you are a vendor then there are a couple more you can use, but for folks in L&D and Training and others who are providing this to their employees, customers, or both OR associations for example – I wouldn’t.

Be Creative

Get them involved. Creativity can work, unless it is so obtuse or weird. Again, you want them to go into the system, and come back and back. Not because it is assigned. Not because it is a certification program. Rather they want to. With e-mails, you will need to know how many times it was opened – there are systems that have that option – email tracking. If the system doesn’t. No matter, use an e-mail tracking program – plenty online. Heck, create a newsletter it goes out once a month, with interesting stuff in there, not just on Chicken U, not just on the courses. Stick in a stakeholder – espousing what you want pushed out. The newsletter works solid with customers – and yes, you can find stakeholders here – cheerleaders. Associations? Ditto. Who would a member rather hear from? You or another member. Get people involved.

Add 3rd party courses/content – but do not indudate. When you only have the content you created – i.e. a variety of topics tied to job roles, or skills, or company-focused; people will go in, but over and over again?

Place in some 3rd party content/courses, then it changes.

How do I do that, without picking courses nobody will want to take?

Send out to your target audience an anonymous survey – this means their responses are anonymous. Here you can zero in on a drawing for those who respond. Pick a variety of topics to choose from. The survey should never be more than one page. You can use statements but be very specific. The never do – “What topics are you interested in? – And you list like 20 topics.

You can go Likert scale here – but make sure it is clear on what one means, two and so on. Stay with a five point likert scale. Let’s say you plan to sell courses to your members.

Bad – “If we offered courses in ABC would you take them? OR “We are looking at offering blah blah, would you be willing to pay for them, if the price did not exceed $75 dollars?” OR you state it, and do not mention the price point.

Good – “We will be offering a course on HR best practices with AI, for $45.” Then go with something like a yes or no option, and always include don’t know. This option should be used carefully because it won’t always apply to what you are trying to achieve or do – i.e. the use case. Want to make money, figure out how much you can charge, rather than just assuming.

FWIW – There is a simple way, beyond using the survey, which I use to identify how much you can charge, to generate revenue, and thus profit with selling content (regardless of the type). And achieve it using Blue Ocean (for example, this is how Apple went when they launched music with the iPod). But there is a strategy for doing so, and an approach. Again, once you know – it is really easy to tap into it, over and over again. What I find is that people always overcharge, and then see limited buying – and justify it, by saying less equals more money.

Bottom Line

I didn’t go heavy into various marketing strategies. You can search and find plenty on the web – without relying on copilot or Gemini to do so. If you have the budget to do so, I recommend using MarketingProfs Find the ones you want to try, and go for it. It will either work, or fall flat. That’s the way it works. Just because X didn’t work, doesn’t mean give up.

What propaganda techiques did I use in this blog? I didn’t use them across the board, but I did use one. Think of it as a test. There isn’t a wrong answer, because who knows – maybe I used more than one.

Thus, ask yourself – why can’t you?

E-Learning 24/7

Latest Insight on AI that is relevant to you

By

onlinefreecourse

-

June 21, 2024

0

Latest Insight on AI that is relevant to you

AI. New Coke. Max Headroom. What do they all have in common? Max was the spokesperson of New Coke. A lot of people think Max was computer generated – a precursor to AI. And AI is left out in the cold, because it actually has nothing to do with any of the above.

Max Headroom was a mega star, okay star in the 80s in the UK. Somehow, he ended up as the spokesperson of New Coke, the same coke which turned out to be a major blunder and removed from shelves in less than three weeks! Max looked like he was computer generated – alas, it was a guy with lots of make-up including prosthetics with a green screen behind him.

Thus, he wasn’t computer generated, nor a pre-cursor to AI. Rather, it was just a guy that landed into the pot of gold playing an annoying character that made the cover of major publications. Yes, a star was born.

I decided to put this post together not to talk about Max Headroom and the actor’s travels thereafter, rather to provide the latest around AI, from a few LLMs, to the carbon footprint (it’s bad), to some hype that hasn’t delivered, studies that might stun you, some surprises around Gen AI, and a variety of additional goodies.

As a promise – any jargon will be reduced to the best of my ability to be explained in examples to everyday life or at least in a laypeople term. A second document – call it my Cheat Sheet to Gen AI (will be published in a couple of weeks and be available on a new site I have for uh, downloading cool content and stuff).

Are you aware?

There is so much out there with AI, that it would require someone to read a lot of publications a night, just to stay current. They would call him a fool – and yes, I am such a fool.

The good news is that you don’t have to – because I will continue to espouse it for you – and how it can apply to L&D and Training at the minimum, and then those pesky enterprises who dove headfirst into AI, without realizing all those interesting nuances.

For some reason, there are people, who think that I can only talk about the negatives, the cons to Gen AI, and lack the deep understanding and insight such as various LLMs, RAG benefits (including what is a RAG), and so forth. For those fine folks, gear up, we are taking off (not on a Boeing though – Ha, I made a funny).

On a side note, there are vendors in our space, that are using RAG – which depending on what you are planning to do within the system with Gen AI – would make sense with the LLM or LLMs they have. Ditto on companies who might have a RAG too. You do not need a RAG, BTW.

Studies – It is the land of confusion

I am amazed, okay not really, on the number of “studies” that have come out around companies’ usage of AI in their budget, impact of financial, usage and so forth. Some contradict each other.

The studies tend to reach out to large numbers of companies, as a whole. The way to look at this, is to recognize that relying solely on one study isn’t the best way to ascertain what is going on with AI.

Study A – Lucidworks – Based on responses from over 2,500 businesses located all over the world

Implementation of AI – 25% have not deployed – they are in a holding pattern – stuck in beta – “like a plane circling an airport for 25 days or more.”
Concerns cited by companies around AI in their businesses – hallucinations, data security and the high cost to run the platforms (foundation is the general term instead of saying LLM, which uh is the foundation)
63% are planning to increase spending on AI in 2024, that is down from 93% who stated they planned to increase spending for 2024
36% plan to keep spending “flat” on AI
And the kicker – 42% have yet to see any financial returns on AI adoption of products

Study B – Slack Workforce Lab

(Polled over 10,000 global desk workers – full-time (30 hrs or more per week – mostly executive, senior, middle and junior management roles), who work with “data, analyze information or this creatively per Workforce Lab’s methodology.”

In other words, global desk workers are a bit misleading – because it’s not Barnaby in the cube next you, unless he was a manager or exec.)

23% usage increase with AI in the workplace (Since Jan 24. The study does not provide a to X date), however, the increase of AI in the workplace since Sept 23 is 60% (again, no info on what date is this up to?)
81% say AI has boosted their productivity and quality of work
93% say that around work-related issues and trustworthiness – have concerns

The study presented additional information on employee engagement. Again, the results are misleading because it is not employees per say, it is specifically folks in management. Anyway, here it is:

18% increase in work-life balance
23% able to manage stress better
24% increase in happiness in the workplace (the study uses the term – overall satisfaction)
29% who state they feel very passionate about their work
73% “of desk workers – again, that isn’t workers, that’s management” believe the hype of AI is warranted
55% of workers in the age bracket (18 to 29) are highly enthusiastic about AI, and the automation to handle parts of their work, only 33% in the age bracket over 60, feel the same way

What the study doesn’t provide is how many of these folks are/were aware of hallucinations (and thus checking before just accepting the info), AI bias just to name two big ones. Another odd thing about the survey is the age bracket 18 to 29, how many of those folks are senior execs or at a minimum of their management level?

Speaking of execs and AI

I tend to find a lot of execs who have no idea about hallucinations – i.e. fake or false information – especially when it is only their content going into the LLM – hence private.

Assumption being if it is only ours, no issues with fake or false information. That is 100% wrong.

Worse, if you decide to remove your content and untrain the LLM, then put new stuff or updates or whatever back in, you will still have remnants of the previous data in the LLM. – Surprise!

Study C – Accenture

Companies that “apply AI – Generative AI) expect to see higher revenue of 25% after five years.”

For me, I would expect a higher revenue than 25% after five years, simply because AI is at a very tiny stage here, and it will only grow – equally though – not all LLMs are the same, and those darn token fees will hit your bottom line – depending on the number of folk using it – let’s say it is the text version only (like GPT-4o – aka GPT4 Omni, which is free for X number of instances, but the paid version – which would be for Enterprise, increases the number.

When GPT-5 rumbles out, again, for Enterprise in this case, you still have token fees).

On the flip side, if you are using an LLM that is multimodal – which 5, and there are others as well – not just OpenAI here – you can do far more than just enter text and get a retort in text.

As an Emeritus Professor of Economics from Oxford told me – “If you are a risk adverse company, I wouldn’t recommend using AI.”

As folks are aware I often talk about the impact of job losses with AI (as it evolves). While there are plenty of people out there who disagree, saying it either won’t be that bad, or it won’t be the case because there will be more jobs.

If you are in the financial services industry

Citigroup says 67% of banking jobs have “high potential” to change or be outsourced by AI. On the flip side though, AI could add 170 billion dollars to the profit “pool” in the banking sector by 2028.

The article adds that they expect no reduction in headcount – once AI related management hiring is identified and resolved.

One item I found of interest from the Citigroup report is their statement that the banking industry will be slow to adopt due to various factors including regulation and the lack of global rules (which is very big).

Study D- Booth School of Business (University of Chicago)

GPT-4 does a better job of analyzing financial data including statements and predications based on those statements
Using “Chain of thought” which directed GPT-4 to identify trends in the financial statements and calculate different financial ratios. They also found that by using a LLM (Large Language Model), GPT-4 achieved an accuracy of 60% compared to humans which is in the low 50% range (relating to predictive accuracy)
Relating to financial acumen GPT-4 produced more profitable trading with higher share ratios (and alpha) beat the stock market.

Did you know?

Sam Altman, CEO of OpenAI, in a discussion around GPT-5, stated that GPT-4 is the “dumbest model any of you will ever have to use again.” Yes, he was comparing it to GPT-5, which is expected to be far superior, nevertheless his reference could have been something like, “GPT-4 was a solid model, but with 5 coming out you can expect superior results, and blah blah.” He didn’t.

The issues with 4 (which Enterprises were using, and maybe still are, without using Omni) included hallucinations and performance – which was awful.

Numerous people found that in the earlier part of the day, the performance and responses were fast; by the end of the day slow or even the “I can’t respond”.

How bad was it? A study found that the model went from 97.6 accuracy on one test in March 2023 to just 2.4 percent by June.

Another issue was the use of DALL-3 whose output went from the ability to save it as a .PNG to now a WebP format – i.e. a web page. The latter requires you to have software that allows you to change the format into PNG.

Open AI was well aware of this early on – after it started to happen – on their own Developer’s forum, people were complaining about it (and made me feel better, because I thought it was something wrong with my computer). As a fun bonus, Open AI still hasn’t fixed the issue.

The Foundational Model

There are two types of foundational models – the one most folks are aware of is LLM – which means large language model. A LLM requires a lot of computing power to such an extent that it’s carbon footprint is very close to the output of a coal plant.

On top of that, to cool down these computers you need water (a vital commodity in 2024 and beyond).

SLM – is a Small Language Model. An SLM will have a smaller set of parameters compared to a LLM, but it can run on a mobile device, even a laptop.

The latter one intrigues me the most when a vendor has a mobile app whereas folks can access content within the app.

Apple has a SLM in their iPhones, even the ability to use the SLM offline. The idea that a vendor in our industry, including learning technology could use a SLM then, seems possible. Phi3 from Microsoft is just one SLM.

The plus of an SLM is the privacy concerns are seriously reduced, they (SLM) can match an LLM in select functions. Even OpenAI recognizes the opportunity for an SLM themselves.

Be aware though that even with an SLM you will still have hallucinations.

Benchmarks

Folks always, okay a lot of people are wondering “how do they know which LLM is better as this, or better performs at that.” Hello, benchmarks.

Here is a short list of the top ones. Always remember though that they base their output on specific items – which they note – like solving a math problem, for example.

Equally remember just for an FYI – that every LLM, even SLM out there, will do things better or worse compared to another. For example, one LLM does a poor job with web links. Another has an issue with summaries.

Benchmarks to Check Out – Click on “Leaderboard” – Some require you to download files, and are complicated. AA, LiveBench, LMSys are easy to use and read.

LiveBench
AGIEval
Hugging Face – Offers various Benchmarks that sit on Hugging Face
LMSys which uses Elo Ratings – Includes GPT 4o (latest LLM from OpenAI)
Artificial Analysis – I’m a fan of this one, because it also shows the latest from Open AI and Anthropic (3.5 Sonnet, matched up to others with performance as one item, fees as another. Best Benchmark out there, IMO.