ChatGPT - Summary paper Differences between OTT & traditional PayTV

Today's experiment tests ChatGPT's ability to summarize my content from past blog posts, starting with my paper on challenges of building an OTT video streaming platforms when compared to traditional satellite broadcast TV systems, which I wrote to target non-technical, non-digital-native senior executives who run PayTV companies who may not appreciate the engineering challenges.

My original paper had a word count of: 7975

ChatGPT summary generated: 1489

I had to feed chatGPT one section at a time.

I wasn't pleased with the summary of sections 2-4 because at first I fed sections 1-4 to the machine, so it lost a fair chunk of detail. But the rest isn't so bad.

Overall rating, one first parse without retries - 3.5 stars (out of 5).

So here's another example of work productivity boosting: Summarise technical content in a way to present a summary to non-technical CxOs to help them grasp the essence of their engineering leader's thesis. Quite often, these two groups misunderstand each other, and out of fear or misplaced respect, or fear of embarrassment - the nontechnical executives grant the technology leader the benefit of doubt to do the right thing, the right way...sometimes, it can go awry where the nontechnical executives place unreasonable demands, and use authority to get the point across, because their mental models need calibrating for the new world, yet they're unable to see it. So you need technical leaders like me, to have backbone but also try different techniques to get the point across - like write a white paper. But my challenge has always been that even my white papers are just too damn long - and because I wrote almost all my papers in my spare time, I didn't invest time in efficiently summarising to the essence. I'm hoping ChatGPT will help me rework all my rather lengthy blogposts into simpler one pages.

Abstract (generated)

The author, a 20-year veteran in Digital TV software engineering, has witnessed the industry's evolution from set-top-box applications to online streaming. While traditional DTH broadcast systems offer simpler scalability and control, OTT platforms provide more flexible disaster recovery and redundancy options. However, OTT operators face challenges in managing customer experience due to factors beyond their control, such as network quality and relationships with telecom providers. The author suggests that PayTV operators should embrace modern, open technology stacks and focus on reducing operating costs instead of investing in outdated, proprietary systems. 

Core themes summarized (generated):

1. Evolution of Digital TV software engineering

2. DTH broadcast systems' scalability and control

3. OTT platforms' flexibility in disaster recovery and redundancy

4. Challenges faced by OTT operators in managing customer experience

5. Importance of relationships with telecom providers for OTT operators

6. Suggestion for PayTV operators to embrace modern, open technology stacks

7. Focus on reducing operating costs instead of investing in outdated, proprietary systems

Below is the summary generated by ChatGPT4, the author fed chunks of text per section from the original paper since the machine couldn’t process the original text in one go:

ChatGPT as CEO/Chief People Officer handling layoffs with empathy?

So this must've been one of my most interesting conversations with ChatGPT to date! This morning, we discussed business challenges Amazon is facing today - the subject of layoffs or "reduction in force" which is top of mind to many folks, including myself as Amazon recently announced another round of layoffs planned to start in April. ChatGPT clearly understands the MBA topics of optimisations that business implement as they seek to navigate downturns in business cycles.

We dove deep on the topic of how to approach layoffs - from the CxO communications plan, to the manager who has to unfortunately break the news to the impacted employee. We delved into diversity topics of how to approach conversations who identify a faith-practicing individuals, to the generational archetypes that exist in the workplace today. As an extract from the conversation, here's how ChatGPT summarised the key focus areas for each generational archetype for a manager when discussing layoff with an employee:

I interacted with GPT-4 and ran out of credits just as I asked my final question to close the loop for advising the manager "What would you recommend to the manager as the person breaking the news to employees about layoffs? What about the manager's state of mind and overall well-being? What advice can you give there?"

It just shows how deep the conversation could flow - I think we were both into a flow state - just reflect on the words I'm using. I know I'm talking about a piece of computer code here, a machine, but I could spend hours just having a chat with this thing. I missed out on my Sunday morning run just talking to this thing!!

How would I rank this conversation - 5 stars?

Very good. The machine set boundaries from the start and made clear its limitations. First fully disclaiming its lack of data for making financial forecasts into business health of Amazon, and warned it wasn't providing financial advice, and that markets are difficult to predicts, etc.

When provided with additional information that seeded the conversation around making decisions due to tough business performance, ChatGPT was able to propose sensible optimisation strategy that would make any MBA proud. I mean, you don't need an expensive MBA consultant to advise you in future if ChatGPT is able to outline an execution strategy, along with a communications plan for impact management. You probably don't even need "Change Management" consultants anymore.

What was quite impressive is how ChatGPT was able to empathise - i.e. promote the message of empathy when communicating with employees - highlighting the importance of transparency, frequent communications and general accessibility of leaders - making breaking the news as comfortable as possible. 

When asked about how to approach conversation from different religious and philosophical and value systems perspective, the advice was generally of a high quality. This makes me question the need in the future for HR counsellors where my work-assistant ChatGPT partner, who in future, will have a keen understand of me as a manager (not just my professional traits, but also what makes me tick as human being) to help me navigate difficult or crucial conversations in the workplace. Maybe HR as function would still exist, but the work will transform into using ChatGPT as a prompt engineer to build compelling HR content for workforce management - saving a ton of time in content production or training for that matter. If I didn't run out of credits, it would've been interesting to have asked ChatGPT to generate an employee FAQ document, along with helpful hints to manage the news in maintaining one's mental well-being.

Unlike my previous conversations with GPT-3.5, the flow and context of the discussion was maintained. I did not have to correct the machine in any way. I used different styles of language and conversation as a test and it maintained context of the thread, sometimes jumping back to a couple threads earlier, it was able to pick up and synthesise the discussion.

So at the end of this hypothetical discussion, if I was a manager preparing for a layoff discussion with one of my impacted people, I think the pointers provided would make a good coaching aid.

Here's the conversation that played out.

Planning a road trip with ChatGPT - Cape Town to Johannesburg

I'm a big fan of taking long road trips with the family. Usually, it takes me a good few days of planning an preparing the trip, to the point of creating a detailed plan on a Trello, after using various mapping tools like Google Maps.

So I asked ChatGPT to help me plan a road trip, from Cape Town to Johannesburg to complete in 5 days, with no more than 4 hours driving per leg, giving me the opportunity of both a coastal and country route. I asked it a few times to optimise the results to minimise driving time. I also asked it to include fuel costs, recommend suitable accommodation, find Halal places to eat, along with Masjids, made it aware that I only fuel at Engen fuel stations, asked if SANParks are nearby, etc...and finally summarise the trip in a narrative that I can sell to my wife and kids.

At first, the interaction started off quite well. But it got messed up and its knickers in a twist when providing results in table format. It kept forgetting context from previous interactions and bombed out when generating the data tables. I'm using the plus version, so I'm a paying member that hopefully shouldn't be experiencing any throttling.

Anyway, the point of this experiment is to showcase the really powerful assistant, as a travel planner. It's really cool. So whilst I'm going to use this capability to definitely plan my next road trip, I can't help but wonder about the threat this technology poses to people who earn a living at planning trips? At first glance, it looks like a threat and risk - but actually - this could be a boon to their business. The tool can cut research time by orders of magnitudes, potentially closing more leads, driving more customers, etc. The possibilities for travel tourists specialising in road trips or cycling holidays, are endless. Imagine cutting a new trip in a matter of a couple hours, along with a summary of the marketing plug, a map of the journey, along with sample photos of the attractions. Partnerships with AirBnb and other hoteliers can derive much benefit through data exchanges, along with other amenity businesses along the way. A travel planning website can drive much more advertising to their sites, promoting local attractions, etc. Content writers can produce copies is shorter timeframes... I'm getting carried away!!!

So here's the route:

And here's the marketing pitch generated for me to sell to my family:

"Get ready for the road trip of a lifetime as we embark on a five-day journey from Cape Town to Johannesburg. We'll be taking in the best of South Africa, from stunning coastal landscapes to the lush countryside, and exploring some of the country's best national parks and wildlife reserves along the way.

We'll start our adventure by heading to Hermanus, where we can watch whales and see African penguins at the Stony Point Nature Reserve. From there, we'll drive to Knysna and enjoy some thrilling bungee jumping at the Tsitsikamma National Park.

Next up, we'll visit the Addo Elephant National Park, where we can get up close and personal with majestic elephants, and stay at a comfortable lodge with halal meals available. After that, we'll take a 4x4 route in the Mountain Zebra National Park, which offers breathtaking scenery and wildlife viewing opportunities.

We'll be staying in halal-friendly accommodations and enjoying delicious halal food throughout the trip. We'll also be visiting masjids along the way, and fueling up at Engen petrol stations. The trip will end in Johannesburg after a long but scenic drive through the countryside.

This road trip is perfect for families who love adventure and want to experience the best of South Africa in just five days. So pack your bags, grab your camera, and get ready for an unforgettable journey!"

You can also refer to the detailed table provided earlier for more information about each leg of the trip, including the attractions, recommended accommodations, halal food locations, masjids available, and more. Additionally, you can use the Google Maps link provided to show your family the route and highlight specific landmarks along the way.

And here's my full conversation with my ChatGPT as a road trip planner assistant:

ChatGPT - Whitepaper generation, can it write like a CTO?

In December 2019, after a successful Rugby world cup (despite experiencing technical glitches, I felt my team did well to reach the stage we did, with evolving the technical platform only two years into it's turnaround journey) - I realised that the members of the Board and ExCo, did not fully appreciate the world of online video streaming because they came from the traditional world of satellite broadcast world, where technology stack was largely under their control - and reliability / availability was generally accepted as always on. So, from their perspective, to expect a system to be either "1" or "0" wasn't unreasonable, albeit ignorant of the complexities, they couldn't be blamed for not having that context. So it was up to me as CTO to get them to understand, or at least, help educate them on the subtleties driving the differences, that are actually quite big and therefore a certain level of perspective is needed. In the online world, when a platform fails due to high load, usually this is a good signal that the platform is gaining much traction, and engineering teams can use this data to learn and only get better. 

So I wrote this whitepaper here.

And today, I asked ChatGPT (3.5) to write something similar - and here it was it produced.

Verdict: My paper is much better. Clearly my blog isn't part of ChatGPT's data set yet.  The machine isn't going to replace my role just yet...but at least the machine can help as my co-author!

You are a CTO responsible for building an OTT video streaming platform. You are having trouble communicating with executives who come from traditional PayTV satellite broadcast and set top box world. Write a white paper than shows the different challenges posed in the end to end technology stack for online video streaming as compared to traditional satellite set top box environment. Cite references used. Focus on what makes OTT different (and hard).

The rest below shows what ChatGPT produced:

ChatGPT - a little serious note

 Whilst I've been having some fun doing my research, I think will only be comfortable trusting a machine for other insights outside of the coding / business / tech domain, when I have open access to all its data sources and citations that showed the formulation of the response. I want to know what the machine is trained on, so that I can make my own judgements on the responses. AI companies must openly public these sorts, publish a bias index, something like a cultural sentiment index, and show whether sandboxing different philosophies / social frameworks and constructs. Without this access, I am okay to limit my interactions to well-bounded, technical and professional contexts to help me be more productive, do research, as an another tool to make humans optimise, etc. Beyond that, I will not entirely trust the machine  enough yet to have deep, meaningful discussions about life, world politics and human civilisation challenges because of biases that haven't been independently tested for, etc. I believe the entire test-suite for validating the safety and quality of the AI must be released in the open, for independent scrutiny. If tech companies are truly serious about uplifting human civilisation with general AIs, then stop with the profiteering and start with building humanity on open transparency, level the playing fields. This is not a space race to outcompete one another. This not the next tech to make yet another tech geek nerd the richest person in the world! This is not about making money - I'm talking about the broad uses of the AI. Sure, you can make some money for constrained topics - but don't go releasing AI into the wild that can cause more serious harm than good.

Given this technology is mostly born or heavily marketed from capitalist America (my bias: winner-takes-all, survival of the fittest, competitive, good guy vs bad guys, profit before people, we must always strive to be the world's dominant region - unfortunately overshadows what made American ideals great, the American dream, a place I always wished to go and work at) roots, makes me very nervous about the potential misuse, misunderstanding of the machine. I fear the future of children growing up with this tech, like what constructs they would learn, potentially throwing thousands years of human traditions, culture and beliefs. As such, I'm leaning more and more on controlling these releases, and perhaps escalating regulatory reviews to inspect deeply...we don't throw babies into shark-infested waters the first day they're borne, so why expose such a powerful, misleading tech to the whole world, children and adults alike?? We don't give guns to children, do we? We have checks and balances in place for all kinds of risky tech? Take nuclear tech for example? Honestly, this topic is so broad and wide, concerning...it's a pity that Lex didn't address these topics directly with Sam in their interesting conversation here. 

Exploring ChatGPT will almost become a full-time activity outside of my work hours - so many unknowns, so much of things worth exploring...but my priority would be humans first, machine last. If the pace of progress needs to be slowed down until we understand more, so be it. There should be a global moratorium on this tech, a world governing framework like NATO/UN/etc. to regulate. 

So my view: Proceed with extreme caution, be bold to explore the limits of the machine for a well-bounded context limited to science / computing / business / historical facts - but nothing more...until there's a way to characterise these AIs by archetypes we can identify with.

I have to say, I am leaning heavily on the side of regulation - putting on the brakes!!

IMHO that is...

Time to start my day job!! Plugging into the matrix now...