Thomas's Innovation Wrap #5

🔬 Cannabis bioengineering, 💊 age reversal, and 💎 avoiding the next big disinformation disaster

Morning,

A few things happened last week! Here’s what I found particularly interesting.

Get the Innovation Wrap in your email each week by signing up here. ✔️

🔬 Biology

Synthetic biology startup Ginkgo Bioworks is profiled in this new 9 minute video (this startup was also mentioned in the Innovation Wrap #1).

Nature writes about the bioengineering of cannabis and how it could enable industrial-scale production of cannabinoids. The article mentions the deal Ginkgo Bioworks did with Cronos Group (a leading cannabis producer) in 2018 where Ginkgo would try to manufacture pure CBD for under $1,000 per kilogram, which compares to the wholesale “natural” price of $5,000 per kilogram.

Michael Gorenstein, Cronos Group’s CEO, presented at the Barclays Global Consumer Staples Conference in Boston last week where he suggested Ginkgo were on track:

The equity is released only for each cannabinoid once we’re able to have a commercial strain that can produce cannabinoids at less than $1,000 per kilo of pure cannabinoid. And I think we’re tracking very positively towards that. And it’s one of the reasons we thought it made sense to actually go and get a manufacturing facility so that we could ferment cannabinoids and be able to hit the timeline when the actual strain organism is ready.

There are opportunities beyond cannabinoids, of course. John Cumbers from SynBioBeta writes in Forbes about how synthetic biology startup funding has exploded, along with five sectors he thinks synthetic biology will soon disrupt.

💎 Artificial Intelligence

A Chinese app called ZAO went viral but also caused privacy concerns - it’s a deepfake app that takes your face from a single photo and superimposes that on various movie characters.

Deepfake videos are more common these days — you can see Will Smith in the Matrix (a role he famously passed on), Jim Carrey in Psycho, and Mel Gibson in Mad Max: Fury Road.

While they’re fun, it’s not hard to imagine deepfakes being used for more nefarious purposes. To hold back the next big disinformation disaster, Facebook, Microsoft, and a number of universities have launched a deepfake detection competition. As part of the challenge, Facebook is creating its own library of deepfakes that participants can use to train their models.

Wired writes about the rise of China’s facial-recognition giants as one of them (Megvii) files for an IPO, while CNBC writes about a growing backlash in China against AI and facial recognition.

🔒 Cybersecurity, Cybercrime, and Cyberwar

Security researchers at Google discovered a major iPhone security flaw earlier this year that let hacked websites access files, messages, and location data from visiting iPhone users. The flaw has already been fixed — Google told Apple about it on February 1 and Apple fixed it on February 7 — but this week, after several months of careful analysis, Google published a deep dive blog post detailing what caused the vulnerability (you can see why they release information like this in their FAQ). Apple responded by attacking Google.

A report last week outlined the first ever recorded cyberattack on the US power grid, though it seems relatively minor. There were no blackouts and no disruption to power generation. Instead, a vulnerability in their firewalls meant hacker(s) could continually reboot them meaning the utility’s control centre lost contact with parts of the grid as devices came off and on the network.

The Economist writes about how AI is changing every aspect of war. The article sorts military applications into three buckets - (1) allowing machines to act without human supervision, (2) processing and interpreting large volumes of data, and (3) aiding or conducting the command and control of war.

Dimitri Simes Jr writing in The American Conservatives argues that AI runs the risk of making our forever wars truly forever.

💊 Health

Is age reversible? A small clinical trial in California with nine volunteers (men aged 51 to 65) who took three common drugs on average shed 2.5 years from their biological ages. This the first time a study has shown a reduction in biological ages in humans, however it’s limited by the fact it was a feasibility study without a placebo. A future trial will involve 100 participants.

“I’d expected to see slowing down of the clock, but not a reversal,” says geneticist Steve Horvath at the University of California, Los Angeles, who conducted the epigenetic analysis. “That felt kind of futuristic.” The findings were published on 5 September in Aging Cell.

Artificially grown human organs may be one step closer with an advancement in 3D printing called SWIFT.

🌞 Renewables and Climate Change

The world is investing less in clean energy as subsidies have declined.

Bill Gates is backing the first high-altitude experiment of one radical climate change solution: solar geoengineering. Solar geoengineering involves creating a massive chemical cloud that mimics the effects of a giant volcanic eruption in order to cool the earth.

⚡ Other Snippets

Is big tech more powerful than some nations? Denmark thinks so, and they’ve now appointed their first foreign ambassador to the technology industry. Meanwhile the US has launched antitrust probes into both Google and Facebook.

Apple’s annual launch event is this Tuesday. Here’s what to expect from The Information, Bloomberg, and Quartz (consensus: all about the iPhone cameras).

Which country will dominate 5G? WSJ writes that China is currently sprinting ahead.

Harvard Business School has a case study on Booking.com, a company where innovation means constant failure.

STEFAN THOMKE: Failure is the status quo. Again, when 9 out of 10 things fail, you’re much more likely to run into a failure than not, so what you have to do is you have to create, basically, an environment that discusses the failures, failures are shared. I always say that not winning is not losing, it’s not the same thing. Not winning means yeah, things fail, but there’s a learning objective here. You can learn from these failures, you share the learning across, and often these failures on input to the next hypothesis, because it’s an iterative process, and you try again, modify the hypothesis, try, and then sometimes these failures lead to some really important insights, which then lead to great experiments, which then lead to big performance improvements.

Stripe, the world’s most valuable private fintech company, is entering the lending business. Credit decisions will be made using “advanced algorithms” that consider trends in payment volume, percentage of repeat customers, and payment frequency rather than traditional FICO stores.

The US is about to become a $10bn recorded music market again (last time 2007) thanks to the growth of music streaming.

The New York Times looks at how high tech is transforming farming (autonomous tractors etc).

Japan will let up to 100 driverless cars roam freely ahead of the 2020 Summer Olympics.

National Geographic produced a video (13 minutes) about how Science Fiction inspires the Future of Science.

Have a great week.

Thomas


About Thomas Rice

Thomas Rice is the portfolio manager for the Perpetual Global Innovation Share Fund, based in Sydney, Australia. You can find him on Twitter at @thomasrice_au.