Hi there,

Uh oh, scientists say El Nino just might threaten the 3 Cs that make life worthwhile: chocolate, cookies and coffee. Fret not, though, we’ll give you a Free Cappuccino for telling a friend about us.

In this Open Letter:

• AI & jobs: Big Tech knows something we don’t
• Cold showers, Zuma-era packages & your chance to make the first page.
• Building ugly: The fastest way to validate and build product demand.
• Free stuff: 50 Startup tools & free cappuccino when you share.

TRENDING NOW

Big Tech knows something about AI we don’t

And it involves… jobs

Noticed how quickly AI optimism turned to doomsday prophets churning out endless lists of the most “at-risk” jobs?

People were speculating just how long it would take AI to start replacing humans, and we shared a short in last Tuesday’s letter about how AI cost 4’000 people their jobs in May alone. So, yeah, the day has come.

If you’ve played around with the likes of Bard, ChatGPT (and GPT-4), Midjourney, and likely most impressively Adobe Firefly with its autofill features, you would’ve been blown away by the capabilities of a technology that seems to have been built overnight (let’s face it not many people were talking about AI around the Christmas table last year).

But AI’s often (hilariously) far from perfect

A lot of companies have taken the plunge and embraced AI. And their results have been… mixed.

In 2018, a self-driving Uber hit and killed a pedestrian (jaywalking) when the car didn’t recognise the person as a pedestrian, (or that pedestrians sometimes jaywalk).

Tech publication CNET used an AI tool (on the very down low) to write a bunch of articles. Of the 77 articles, more than half needed corrections for incorrect information and “replacing phrases that were not entirely original” (read: probably plagiarised).

Google/Alphabet’s now infamous first public demo of Bard incorrectly attributed the 1st pic of a planet outside our Solar System to the James Webb Space Telescope.

And to have a good chuckle, check how impressively ChatGPT just creates research papers out of thin air when we asked it to write a piece about emerging trends. Masterfully done, lol.

And that’s just the tip of the iceberg.

Which begs the question: is AI really a threat to jobs? Or will everyone just get rehired in a few months’ time when we realise accuracy is actually important?

That said, though, there are some signs that Big Tech knows something we don’t…

Betting on the automation revolution

There’re, quite frankly, some big companies that shouldn’t exist but are being kept alive as if waiting for AI to become truly viable.

Take E-commerce, for example. Amazon’s retail is yet to make meaningful profit and Takealot is still not profitable. So why keep them going?

One plausible reason could be that once AI and robotics become mainstream – think fully automated warehouses with self-driving car deliveries – their costs will go down substantially, and they’ll be positioned to dominate.

The same applies to Uber which has been burning money for years. But it could become profitable if you replace human drivers with automatons. Not sure how minibus taxis that already have a tumultuous relationship with ride-hailing services will respond to this, but hey, it’s coming.

In fact, it's probably worthwhile keeping an eye out for tech companies that will benefit substantially from cost savings once the inevitable robot workers take their place next to us at work. Why? Well, those jobs are likely getting swapped out and what’s more, backing some of these public companies could bring forth some handsome returns (not financial advice, of course, DYOR).

But What Jobs Will Remain?

Rest assured, though, there will always be work.

Because that’s what capitalism is – wage-earning workers are also the buyers of goods and services, which is what creates taxable events. You can’t have one without the other. Remove any of those from the equation and the world ends – you know, robot-domination-style like The Matrix.

That’s why we always say the advent of AI will likely mean entirely new job types we haven’t even dreamt of yet and certain jobs becoming more prominent.

So where will we get money to keep consuming?

Hard to say exactly. In fact when we were at university 20 years ago, none of the things we do today existed. Well, at least not in the form we are doing it now.

But here are some predictions of jobs that might end up paying more in the future:

• The most obvious “new” job is an AI Prompt Engineer. Are you savvy at getting AI to respond in useful ways? Then you might just land a 6-figure job. As corporations seek to incorporate AI into their workflows, engineers who understand how to make it useful, will save companies millions and thus justify high salaries.
• Authentic communities that foster human connection – churches, community groups, hobby groups and perhaps even political movements.
• Theatre and live performance: TV killed theatre for the mainstream guy – it just scaled better and was more convenient. But now with AI becoming so powerful that entire feature-length films can be created without real people, perhaps the appeal of TV will drop and a desire for more authentic human products will rise. With AI-created film becoming indistinguishable from real people, the only way to get that would be in theatre. The same applies to live musical performances by humans playing real instruments.
• Finally, perhaps we will get some income from the money bots make and don’t have to work. We covered this in one of our favourite Open Letters about UBI and Worldcoin.

What do you think? Are jobs going to disappear? Hit reply and let us know what you think the future of work will look like.

IN SHORT

🚿 Government to give you a cold shower. Under the guise of power management, the government wants to roll out smart meters to remotely manage our geysers’ electricity usage. What could be an effective way to manage electricity usage, could also be used to give political opponents a cold shower.

🔻 Dropcoin. SOL, ADA and Matic slide as SEC’s crackdown on crypto exchanges forces Robinhood to delist some alts from its platform causing massive sell pressure.

📫 SAPO Delivers. Eventually. In May we made a meme about the South African Post Office hard at work delivering Christmas prezzies from 2017. Turns out we weren’t wrong. A package sent from NYC that got lost by SAPO finally arrived in Durban last week – nearly 13 years after it was sent.

🧠 Next Level MedTech. A new brain-spine interface device that decodes the signals in your brain associated with movement can help spinal injury patients walk (and stand) again.

🐛 Skynet for pests. The team at Carnegie-Melon built an insane robot to control the spread of invasive species. Basically an all-electric tractor with a robotic arm and computer vision hunts and destroys bug eggs.

🔥 Your moment to shine. With over 87% of subreddits down as mods protest Reddit’s proposed plans to charge third-party apps for accessing its API, the next 48 hours is probably the best chance your memes will ever have of reaching the first page – post boldly!

­

BUILDER’S CORNER

The “Ugly” Route to Better Products

Building a product is a journey full of twists, turns, and sometimes, roadblocks. The excitement of seeing an idea materialize can often lead to premature launches without necessary market validation.

One of the lessons learned from seasoned entrepreneurs is that sometimes the route to validation is less polished than we think. In the startup world, we call it the “ugly” approach, and here's why you might want to consider it.

Don’t @ us UI designers

Let's look at a story that surfaced on Reddit. An entrepreneur had a bright idea for an SEO keyword-searching tool. Like many of us, he invested a lot of time and resources into creating a flawless product with refined architecture and intuitive UX/UI. But when he showcased his product to communities, people were confused, and not a single soul signed up. This lack of response forced him to abandon his product.

But he didn't give up. He had another idea. This time, he decided to flip the traditional development process. Instead of refining the product first, he tested its core functionality with the audience. He created a basic, “ugly” video with no music, audio, or subtitles, and posted it on Reddit. This simple, raw video clearly displayed the tool in action on his laptop.

The response was overwhelmingly positive, with people even demanding the tool! He hastily set up a 'coming soon' page to capture emails, amassing 36 registrations and a promise of many more from Reddit. This new approach led him to what seemed like a genuine product-market fit. You can check out his Reddit post here and the 'coming soon' page here.

So, how can you apply this strategy? Here's a step-by-step guide:

1. Embrace the Ugly: Put your product's core functionality front and centre. Forget about the frills, and focus on the problem you're trying to solve. Remember, it doesn't have to be perfect – it just needs to work.

2. Show, Don't Tell: Demonstrate how your product works in the simplest way possible. If people can understand its value through a straightforward, “ugly” video, you know you're on the right track.

3. Get Feedback Early: Don't wait until your product is fully baked. Share it with your target audience and learn from their reactions. Their feedback is essential in aligning your product with market demand.

4. Be Ready for Interest: Prepare a simple way to capture interest. When your audience resonates with your product, make sure you can hold onto them, be it through a simple website or an email signup sheet.

5. Evolve with Your Users: Take your users on your product development journey. Their needs and demands will guide you towards a truly valuable product.

In the end, remember that a little “ugly” never hurt anyone. In fact, it may lead to a better, more valuable product because it emphasizes function over form.

So, founders, dare to embrace the “ugly” approach for your next product validation. You might just hit the product-market fit jackpot.

Busy with an “ugly” product? Hit reply and let us know, we’d be keen to check it out.

DID YOU LIKE THIS WEEK’S OPEN LETTER?

Did you enjoy this Open Letter?

• ❤️ Loved it
• 😔 You've done better
• 💩 Today was meh

Login or Subscribe to participate in polls.

TELL YOUR FRIENDS

This Open Letter is brought to you by Renier Kriel, Jason Mill, Elvorne Palmer and Bobby Sequeira.

Are you social? Follow our brand new Twitter, IG, Facebook or LinkedIn Page managed by Mia Visser.

Did we miss something? Hit reply and tell us what trends you’d like us to explore next.

Did someone forward you this email? Sign up here.

Find more awesome business ideas from South Africa's favourite startup and tech newsletter.