Lazy Sunday #14 - brain chips, angry shreds and African caviar

Morning, afternoon, evening!

The brain requires approximately a million times less energy to perform than a computer. So let’s just say, it is somewhat more energy efficient.

Would it make sense to grow brain tissue into neural networks on top of tiny chips, and thus create a more energy-efficient computing model?

The team at Cortical Labs believe exactly that and got invited to Jeff Bezos’ “Mars 2024” conference. But more on that below.

We also talk about shredding angry thoughts (literally!), how we downplay risk, group decision making and African caviar. 🐟️ 

Find my research: neuroscience

An intro to Australia-based Cortical Labs - from Jeff Bezos’ Mars 2024 conference

This is what Thomas Oxley, CEO of Synchron, who attended as well, said about the conference:

On we go to Hon Weng’s presentation. It is mind-blowing to see what’s cooking (yes at times it’s a little bit nerdy, but it is not overly medical.)

Unfortunately I can not embed the video properly. But please do have a look or bookmark for later. It is fascinating!!

(Youtube)

Angry? Shred it!

These are the kind of studies highlighting life hacks nobody will tell you…well unless you read studies or this newsletter. 🙃

Researchers in Japan suggest that writing down angry thoughts with pen and paper, e.g. after a fight with a colleague or partner, and then throwing it away or shred the paper reduces anger significantly.

Apparently, writing it down and leaving it sit on the desk does not have the same effect.

I mean, how easy is this? Let me know when you tried it.

(Nature)

The riskier the move the more we downplay the risk

A 2019 study from the Netherlands suggests that we human beings are funny sometimes.

No kidding!

The bigger the risk and uncertainty of a certain situation, the more we tend to downplay the potential chances of losses.

We engage in “wishful thinking”.

Oh, the early-stage investment will be fine!

Oh, climate change won’t have severe effects on my region!

Oh, my co-founder and me, we are cool (despite the fact we have been continuously fighting for the last 3 months).

On the other side, maybe this wishful thinking is the driver for many solutions to problems that seemed unsolvable at first.

There is a second paper from the team, confirming the main outcomes. I haven’t read it yet though (hi, paywalls!)

(Tinbergen Institute)

Find my research: office

Group thinking is suboptimal if you want the best results

I have always held the belief that group-driven decision-making is not always the most successful.

The risk profile of the individuals converge to the average group risk profile of the group if a group decision is to be made. Outliers are averaged.

Sure, it helps to keep the spirits high (”we are a good team”). For some decisions, this is the way to go, e.g. when risks are to be minimized. This is not always the preferred outcome.

There are a couple of methods to get decisions through a group, many of them laid down in the article. My favourites:

• Allow for alternative testing - allow placing small bets through alternative processes. Often the scarcity of full support makes teams especially good at thinking around the corner.

• Consensus minus X rule - cultivate some degree of opposition by allowing for a decision with sceptic a minority

• Devil’s advocate - let somebody ask all the hard questions. All of them, without being judged and labelled. Fun game.

• Ambitious timelines - this is so difficult because what is ambitious? Ambitious is when the goal and timelines are almost painfully hard. Go for them! If you are serious about the problem, you will find a way to execute.

• Disagreement over Consensus: it’s ok to not have everyone agree, but everyone needs to run with the decision to make the most out of it

• Juniors speak first - sometimes the most junior has the best ideas, but when seniors speak first it can skew the view of more junior people on the team - or have them not speak up at all.

The last two are strong favorites of my dear friend Jeff Bezos.

The article also discussed why corporate venture arms aren’t successful: they are too slow!

(HBR)

On top of tech

A short journey with Neurable

Earlier this year, I mentioned the Neurable headphones, which track brainwaves and tell the listener when to focus or hit the pause button at work.

This is an interesting article about the story of the company, their goal and strategies both from product development and go-to-market. Worth a read for the curious!

(MGMTBoston)

Playing BCI games without the calibration

emember when you got a new joystick for your flight simulator, and you needed to calibrate it first? A lot of the time it did not work immediately, it had delays etc.

Each brain has individual brainwave patterns and noise signature. When people use current versions of non-invasive brain interface computers (BCI), the devices require longer calibration to “tune into” your brain. This kinda takes away a lot of the fun because you can’t just plug it in and get going for optimal results.

A new BCI model allows you to play games (ok, simple ones, but still games) without the time-consuming calibration. A new “decoder”, an ML model, which can analyse and, abstract incoming data makes it usable after a very short period of training on the individual.

If BCI wants to go mainstream, the calibration phase needs to be as short as possible. Currently, BCI like Neuralink require weeks and months. Even Cochlear implants require a couple of weeks for the brain to get used to.

(PNAS Nexus Research Paper / Article)

UAE and Saudi Arabia invest heavily in tech

In the 2000s and early 2010s, Russia spent an enormous amount of money on tech infrastructure, funding of startups and business schools. I don’t think it was a surprise, that the current CEO of Techstars, Maelle Gavet, used to lead the biggest e-commerce company in Russia, ozon.ru. Russia had to shift from oil dependency.

Technology and related services don’t need the space, investment and natural resources.

Similar shifts are vivid in the Middle East, with Qatar, UAE and Saudi Arabia leading with the scale of investments that is 🤯.

A $100 billion USD from Abu Dhabi alone to be invested in AI, both infrastructure / hardware and software. Let me compare that to Germany, where the government thinks €1.75 billion for 2024 is sufficient.

For the sake of comparison, let me assume Germany sticks to €1.75 billion until 2030, a total of €12.25 billion.

The difference in per capita investment is immense! Almost 70x!

Does somebody have data for e.g. India or Australia?

(Bloomberg)

Cyborgs and Centaurs

It is a slightly older study from 2023. Still relevant.

The authors conducted two randomized field experiments with 758 Boston Consulting Group consultants, assigning them tasks either within or outside the current capabilities range of AI.

For tasks inside their range, consultants using AI were significantly more productive and produced higher quality results compared to the control group.

For a task designed to be outside their experience, consultants using AI were less likely to produce correct solutions.

Two emerging human-AI collaboration models show promise for navigating AI's jagged capabilities frontier: “Centaurs” strategically delegating sub-tasks, and “Cyborgs” closely integrating their efforts with AI.

(HBS)

Media Box

Cars collect driving data to sell to insurances

“Data is the New Oil of the Digital Economy” was the subject line of a Wired article from 2014.

Some companies took this serious. They buried paragraphs deep within their T&C’s that allowed them to collect driver data (mileage, acceleration, speed etc.) and sell the data to insurance companies. The insurance companies used the data to create individual risk profiles of drivers, which then impacted insurance premiums.

I don’t mind anonymized data sharing as long as it has full, transparent consent from the individual.

Data is helpful to improve services, think about health care, education, improved traffic in cities. But if it is used to further exploit people to squeeze out yet another dollar without these people knowing that their data is being shared, I am getting queasy.

Read the fine print folks!

(NYTimes)

Misc but not least…

I will keep trying to bring the unexpected in this section.

So how about we talk caviar this week?

Caviar from Africa? From Madagascar, to be more specific?

Seems to be a hit, considering Michelin star chefs put it on their plates.

As always thank you for reading.

Please share the link on social media or by forwarding this episode of Lazy Sunday to your friends and colleagues.

Hit to subscribe.

Alex