Adverse training AI models: a big self-destruct button?

21

October

2023

No ratings yet.

“Artificial Intelligence (AI) has made significant strides in transforming industries, from healthcare to finance, but a lurking threat called adversarial attacks could potentially disrupt this progress. Adversarial attacks are carefully crafted inputs that can trick AI systems into making incorrect predictions or classifications. Here’s why they pose a formidable challenge to the AI industry.”

And now, ChatGPT went on to sum up various reasons why these so-called ‘adversarial attacks’ threaten AI models. Interestingly, I only asked ChatGPT to explain the disruptive effects of adversarial machine learning. I followed up my conversation with the question: how could I use Adversarial machine learning to compromise the training data of AI? Evidently, the answer I got was: “I can’t help you with that”. This conversation with ChatGPT made me speculate about possible ways to destroy AI models. Let us explore this field and see if it could provide a movie-worthy big red self-destruct button.

The Gibbon: a textbook example

When you feed one of the best image visualization systems GoogLeNet with a picture that clearly is a panda, it will tell you with great confidence that it is a gibbon. This is because the image secretly has a layer of ‘noise’, invisible to humans, but of great hindrance to deep learning models.

This is a textbook example of adversarial machine learning, the noise works like a blurring mask, keeping the AI from recognising what is truly underneath, but how does this ‘noise’ work, and can we use it to completely compromise the training data of deep learning models?

Deep neural networks and the loss function

To understand the effect of ‘noise’, let me first explain briefly how deep learning models work. Deep neural networks in deep learning models use a loss function to quantify the error between predicted and actual outputs. During training, the network aims to minimize this loss. Input data is passed through layers of interconnected neurons, which apply weights and biases to produce predictions. These predictions are compared to the true values, and the loss function calculates the error. Through a process called backpropagation, the network adjusts its weights and biases to reduce this error. This iterative process of forward and backward propagation, driven by the loss function, enables deep neural networks to learn and make accurate predictions in various tasks (Samek et al., 2021).

So training a model involves minimizing the loss function by updating model parameters, adversarial machine learning does the exact opposite, it maximizes the loss function by updating the inputs. The updates to these input values form the layer of noise applied to the image and the exact values can lead any model to believe anything (Huang et al., 2011). But can this practice be used to compromise entire models? Or is it just a ‘party trick’?

Adversarial attacks

Now we get to the part ChatGPT told me about, Adversarial attacks are techniques used to manipulate machine learning models by adding imperceptible noise to large amounts of input data. Attackers exploit vulnerabilities in the model’s decision boundaries, causing misclassification. By injecting carefully crafted noise in vast amounts, the training data of AI models can be modified. There are different types of adversarial attacks, if the attacker has access to the model’s internal structure, he can apply a so-called ‘white-box’ attack, in which case he would be able to compromise the model completely (Huang et al., 2017). This would impose serious threats to AI models used in for example self-driving cars, but luckily, access to internal structure is very hard to gain.

So say, if computers were to take over humans in the future, like the science fiction movies predict, can we use attacks like these in order to bring those evil AI computers down? Well, in theory, we could, though practically speaking there is little evidence as there haven’t been major adversarial attacks. Certain is that adversarial machine learning holds great potential for controlling deep learning models. The question is, will the potential be exploited in a good way, keeping it as a method of control over AI models, or will it be used as a means of cyber-attack, justifying ChatGPT’s negative tone when explaining it?

References

Huang, L., Joseph, A. D., Nelson, B., Rubinstein, B. I., & Tygar, J. D. (2011, October). Adversarial machine learning. In Proceedings of the 4th ACM workshop on Security and artificial intelligence (pp. 43-58).

Huang, S., Papernot, N., Goodfellow, I., Duan, Y., & Abbeel, P. (2017). Adversarial attacks on neural network policies. arXiv preprint arXiv:1702.02284.

Samek, W., Montavon, G., Lapuschkin, S., Anders, C. J., & Müller, K. R. (2021). Explaining deep neural networks and beyond: A review of methods and applications. Proceedings of the IEEE109(3), 247-278.

Please rate this

AI-Powered Learning: My Adventure with TutorAI

16

October

2023

No ratings yet.

Subscribe to continue reading

Subscribe to get access to the rest of this post and other subscriber-only content.

Please rate this

Weapons of mass destruction – why Uncle Sam wants you.

14

October

2023

No ratings yet.

The Second World War was the cradle for national and geopolitical informational wars, with both sides firing rapid rounds of propaganda at each other. Because of the lack of connectivity (internet), simple pamphlets had the power to plant theories in entire civilizations. In today’s digital age, where everything and everyone is connected, the influence of artificial intelligence on political propaganda cannot be underestimated. This raises concern as, unlike in the Second World War, the informational wars being fought today extend themselves to national politics in almost every first-world country.

Let us take a look at the world’s most popular political battlefield; the US elections; in 2016, a bunch of tweets containing false claims led to a shooting in a pizza shop (NOS, 2016), these tweets had no research backing the information they were transmitting, but fired at the right audience they had significant power. Individuals have immediate access to (mis)information, this is a major opportunity for political powers wanting to gain support by polarising their battlefield.

Probably nothing that I have said to this point is new to you, so shouldn’t you just stop reading this blog and switch to social media to give your dopamine levels a boost? If you were to do that, misinformation would come your way six times faster than truthful information, and you contribute to this lovely statistic (Langin, 2018). This is exactly the essence of the matter, as it is estimated that by 2026, 90% of social media will be AI-generated (Facing reality?, 2022). Combine the presence of AI in social media with the power of fake news, bundle these in propaganda, and add to that a grim conflict like the ones taking place in East Europe or the Middle East right now, and you have got yourself the modern-day weapon of mass destruction, congratulations! But of course, you have got no business in all this so why bother to interfere, well, there is a big chance that you will share misinformation yourself when transmitting information online (Fake news shared on social media U.S. | Statista, 2023). Whether you want it or not, Uncle Sam already has you, and you will be part of the problem.

Artificial intelligence is about to play a significant role in geopolitics and in times of war the power of artificial intelligence is even greater, luckily full potential of these powers hasn’t been reached yet, but it is inevitable that this will happen soon. Therefore, it is essential that we open the discussion not about preventing the use of artificial intelligence in creating conflict and polarising civilisations, but about the use of artificial intelligence to repair the damages it does; to counterattack the false information it is able to generate, to solve conflicts it helps create, and to unite groups of people it divides initially. What is the best way for us to not be part of the problem but part of the solution?

References

Facing reality?: Law Enforcement and the Challenge of Deepfakes : an Observatory Report from the Europol Innovation Lab. (2022).

Fake news shared on social media U.S. | Statista. (2023, 21 maart). Statista. https://www.statista.com/statistics/657111/fake-news-sharing-online/

Langin, K. (2018). Fake news spreads faster than true news on Twitter—thanks to people, not bots. Science. https://doi.org/10.1126/science.aat5350

NOS. (2016, 5 december). Nepnieuws leidt tot schietpartij in restaurant VS. NOS. https://nos.nl/artikel/2146586-nepnieuws-leidt-tot-schietpartij-in-restaurant-vs

Please rate this

0 to 60 mph in 1.9 seconds: The Tesla Roadster.

8

October

2020

The pinnacle of electrification of cars.
As the successor to the first production car of Tesla, which was the 2008 Roadster, the development of the new Tesla Roadster was announced by Tesla CEO Elon Musk in November 2017.

TeslaRoadster2

The fully electric vehicle is said to be released after the release of the renewed Model S, currently Tesla’s most famous model car. Tesla promises a 0-60 of 1.9 seconds with a top speed over 250 mph (400 km/h). The Roadster would be capable of such incredible performance figures due to its staggering 10,000 Nm of torque and all-wheel drive system. This would make the Tesla Roadster the fastest car in the world.

The Roadster would break all records for acceleration and performance compared to traditional super cars with combustion engines. With an expected range of 1,000 km, the range for electric vehicles would be greatly outperformed. Currently, this record is also held by Tesla, with the Tesla Model Y which has a range of 508 km. This is the most interesting point to me. Although the Tesla Roadster might look like an electric toy for rich people, in reality, I think the Tesla Roadster will achieve 2 things that are very important in our search for a sustainable future.

The 2 reasons:
1. Just like with their Model S, the Tesla Roadster will make electric vehicles more appealing. Before the introduction of the Model S, electric vehicles were mostly low performance cars with boring designs. The segment was mainly intended for early adopters: drivers with a strong interest in sustainability and wanting to compromise on performance and design, in return for a more eco-friendly footprint with regards to their driving. After the Model S took the market by storm, the image of electric vehicles was completely changed. No longer where electric cars associated with compromising performance and boring designs. Instead, Tesla made electric vehicles a reasonable choice in the executive segment. The Tesla Roadster is capable of doing the same. Outperforming “classic” super cars, the Roadster will increase the appeal of electric driving world wide.

Elon happy

2. The tesla Roadster will push electric vehicle technology further with record breaking acceleration, top speed and most importantly: range. Currently, electric vehicles are known for their acceleration. The electric drive train makes it possible for the cars to have full access to their potential power from the moment you hit the pedal. However, their topspeed and range are often limited, due to the battery size. Batteries are heavy and therefore companies have to find a balance between the required performance (speed, acceleration, range) and how heavy they want the car to be. After all, the heavier the car, the more the weight is influencing the desired performance. I think the Tesla Roadster will push other car manufacturers to further develop the electrification of cars. This will result in more widely available models with increased performance at a more consume friendly price.

0to100realquick2

Do you have some savings laying around and has this blog article made you interested in the Tesla Roadster?
Prices are still to be announced for the European market, but the base model is expected to cost 200,000 dollar in the US, but the first 1,000 production cars (announced as the Founder series) will be priced at 250,000 dollar in the US. Future customers can pre-order the Roadster with a base reservation of 43.000 euro and a founders-serie reservation of 215,000 euro (for the Netherlands). For more information, check out Tesla’s  website:  https://www.tesla.com/nl_NL/roadster?redirect=no

5/5 (5)

Please rate this

BIM, Meet Gertrude!

6

October

2020

Gertrude enjoying a well deserved drink during her performance. 

In August 2020, famous tech entrepreneur Elon Musk revealed his latest technological project: a pig called Gertrude. On first sight, Gertrude looks like an ordinary Pig. She seems healthy, curious, and eager to taste some delicious snacks. When looking at her, it is hard to imagine how she managed to get one of the world’s most radical and well known tech entrepreneurs so excited. Gertrude just seems normal.

This is exactly the point!

ElonMuskGotcha

Elon Musk “Gotcha”

Gertrude is no ordinary pig. She has been surgically implanted with a brain-monitoring chip, Link V0.9, created by one of Elon Musk’s latest start-ups named Neuralink.

Neuralink was founded in 2016, by Elon Musk and several neuroscientists. The short term goal of the company is to create devices to treat serious brain diseases and overcome damaged nervous systems. Our brain is made up of 86 billion neurons: nerve cells which send and receive information through electrical signals. According to Neuralink, your brain is like electric wiring. Rather than having neurons send electrical signals, these signals could be send and received by a wireless Neuralink chip.

To simplify: Link is a Fitbit in your skull with tiny wires

The presentation in August was intended to display that the current version of the Link chip works and has no visible side-effects for its user. The user, in this case Gertrude, behaves and acts like she would without it. The chip is designed to be planted directly into the brain by a surgical robot. Getting a Link would be a same day surgery which could take less than an hour. This creates opportunities for Neuralink to go to the next stage: the first human implantation. Elon Musk expressed that the company is preparing for this step, which will take place after further safety testing and receiving the required approvals.

The long term goal of the Neuralink is even more ambitious: human enhancement through merging the human brain with AI. The system could help people store memories, or download their mind into robotic bodies. An almost science-fictional idea, fuelled by Elon Musk’s fear of Artificial Intelligence (AI). Already in 2014, Musk called AI “the biggest existential threat to humanity”. He fears, that with the current development rate, AI will soon reach the singularity: the point where AI has reached intelligence levels substantially greater than that of the human brain and technological growth has become uncontrollable and irreversible, causing unforeseeable effects to human civilization. Hollywood has given us examples of this with The Matrix and Terminator. With the strategy of “if you cannot beat them, join them”, Elon Musk sees the innovation done by Neuralink as an answer to this (hypothetical) catastrophical point in time. By allowing human brains to merge with AI, Elon Musk wants to vastly increase the capabilities of humankind and prevent human extinction.

Singularity
Man versus Machine

So, will we all soon have Link like chips in our brains while we await the AI-apocalypse?

Probably not. Currently, the Link V0.9 only covers data collected from a small number of neurons in a coin size part of the cortex. With regards to Gertrude, Neuralink’s pig whom we met earlier in this article, this means being able to wirelessly monitor her brain activity in a part of the brain linked to the nerves in her snout. When Gertrude’s snout is touched, the Neuralink system can registers the neural spikes produced by the neurons firing electronical signals. However, in contrast: major human functions typically involve millions of neurons from different parts of the brain. To make the device capable of helping patients with brain diseases or damaged nervous system, it will need to become capable of collecting larger quantities of data from multiple different areas in the brain.

On top of that, brain research has not yet achieved a complete understanding of the human brain. There are many functions and connections that are not yet understood. It appears that the ambitions of both Elon Musk and Neuralink are ahead of current scientific understanding.

So, what next?

Neuralink has received a Breakthrough Device Designation from the US Food and Drug Administration (FDA), the organisation that regulates the quality of medical products. This means Neuralink has the opportunity to interact with FDA’s experts during the premarket development phase and opens the opportunity towards human testing. The first clinical trials will be done on a small group of patients with severe spinal cord injuries, to see if they can regain motor functions through thoughts alone. For now a medical goal with potentially life changing outcomes, while we wait for science to catch up with Elon Musk’s ambitions.

 Neuralink-Logo

Thank you for reading. Did this article spark your interest?
For more information, I recommend you to check out Neuralink’s website https://neuralink.com/

Curious how Gertrude is doing?
Neuralink often posts updates on their Instagram page https://www.instagram.com/neura.link/?hl=en

Want to read more BIM-articles like this?
Check out relating articles created by other BIM-students in 2020:

Sources used for this article:

4.88/5 (8)

Please rate this

Professors! Get online or get out!

16

October

2019

5/5 (1)

Please rate this

As a BIM master student, I was quite surprised when I heard that none of the courses were recorded and therefore available online. Everyone I ever spoke about it was enthusiastic about recorded lectures. Maybe all of my friends are just lazy students (like me), who prefer to stay in bed rather than going to a 9 am lecture, but I genuinly think it offers more convenience than it has disadvantages. Me wondering this was the main reason for me to write on this subject.


MOOC stands for Massive Open Online Courses, and are (often free) courses that are available to the public through online lectures and assignments (EdX, 2019). It provides great advantages as you can enroll from anywhere around the world, as long as you have access to a decent internet connection.

First of all, and maybe the most obvious advantage of MOOC’s, it that the internet knows no borders. Of course we all know the Great Chinese Firewall, but someone from South-Korea is able to enter a website from a Colombian local bee farm. Therefore, people from more abandoned areas, like sub-Saharan Africa are able to enter these courses as long as there is a decent internet connection and a streaming device. According to UNESCO (2016), sub-Saharan Africa has the highest rates of education exclusion in the world. Almost 60% of all youth between 15 and 17 there are not in school. Yes, they still require a streaming device, but a phone screen is in theory enough, and video projectors can be installed in classrooms.

This brings us to another advantage of MOOC’s, there is (in theory) no maximum student capacity. As it is a digital product, it can in theory be copied infinitely without reducing in quality. This means an enormous amount of people could follow the course of a single professor. This seems like a situation that only has benefits, but there are some risks. If a single professor is enough to educate a massive group of people, then I foresee a decrease of the need for professors. This may lead to many professors losing their job, and having to seek other ways to earn a living.

MOOC’s being a digital good also brings a major risk, the risk of the course content being copied and spread without consent and compensation. Screens can be recorded and assignments being copied. Websites like The PirateBay that provide a lot of illegal content are nowadays still available, whether it is through a proxy server or not). A solution must be sought to prevent piracy, because a single pirate is enough to create a lot of damage.

 

Another advantage of MOOC’s is that it provides an opportunity to gather data about its students. It can be tracked how much and when students spend time on the website, and which classes and courses are more and less attractive. Students may be able to provide a rating and a comment after every course. A risk of having too many students enrolled, is that a single professor may not be able to answer all questions or analyze feedback. This proves that a MOOC is not simply a professor with a webcam, but really requires a well-structured team or organization.

I would advise professors and universities to brainstorm about threats and opportunities in the increasingly digitized society. I believe that it’s very important not to miss the boat and to exploit first-mover advantages. Otherwise, you will remain the incumbent, while others become the disruptors.

 

References

EdX. (2019). mooc.org. Retrieved October 16, 2019, from http://mooc.org/.

UNESCO. (2016). Education in Africa. Retrieved October 16, 2019, from http://uis.unesco.org/en/topic/education-africa.

 

5/5 (2) The Threat of Deepfakes

12

October

2019

Please rate this

Last summer an app called DeepNude caused a lot of controversy in the (social) media. Deepnude was an AI based piece of software with the ability to create a very realistic nude pictures of any uploaded face in the app. Mass criticism followed, the app’s servers got overloaded by curious people and not much later, the app went offline permanently. Deepnude stated on twitter that the probability is misuse was too high and that the world “was not ready yet”. The app never came back online ever since  (Deepnude Twitter, 2019). It shows that deepfake technology is becoming available to the public sooner than we thought, including all potential risks.

A definition for DeepFake is “AI-based technology used to produce or alter video content so that it presents something that didn’t, in fact, occur” (Rouse, 2019). As deepfake is AI-based technology it is able to improve over time, as the amount of data input increases and the technology learns to how to create better output. In my opinion deepfake has an amazing potential in the entertainment industry, but there is a serious risk when the technology gets misused. The AI technology makes it harder and harder for humans to distinguish real videos from fake ones. Deepfake videos of world-leaders like Trump and Putin are already to be found on the internet. Also deepfake porn videos of celebrities are being discovered once in a while.

With the upcoming presidential elections of 2020 in the United States, politicians and and many others are seeking solutions to prevent a similar scenario like the 2017 elections. The 2017 presidential elections were characterized by the spread of fake news and the ongoing allegations resulting from it. These events very likely influenced the outcome of those elections (CNN, 2019). Recently the state of California passed a law which “criminalizes the creation and distribution of video content (as well as still images and audio) that are faked to pass off as genuine footage of politicians. (Winder, 2019).” In 2020 we’ll find out whether deepfakes have been restricted succesfully.

I hope developers and users of deepfake technology will become aware of the huge threats of deepfake, and will use it in a responsible way. It is also important for society to stay critical at their news sources and that they prevent supporting these types of technology misuse. According to Wired (Knight, 2019), Google has released thousands of deepfake videos to be used as AI input to detect other deepfake videos. Another company called Deeptrace is using deep learning and AI in order to detect and monitor deepfake videos (Deeptrace, sd).

See you in 2020…

References

CNN. (2019). 2016 Presidential Election Investigation Fast Facts. Retrieved from CNN: https://edition.cnn.com/2017/10/12/us/2016-presidential-election-investigation-fast-facts/index.html

Deepnude Twitter. (2019). deepnudeapp Twitter. Retrieved from Twitter: https://twitter.com/deepnudeapp

Deeptrace. (n.d.). About Deeptrace. Retrieved from Deeptrace: https://deeptracelabs.com/about/

Knight, W. (2019). Even the AI Behind Deepfakes Can’t Save Us From Being Duped. Retrieved from Wired: https://www.wired.com/story/ai-deepfakes-cant-save-us-duped/

Rouse, M. (2019). What is deepfake (deep fake AI). Retrieved from TechTarget: https://whatis.techtarget.com/definition/deepfake

Winder, D. (2019). Forget Fake News, Deepfake Videos Are Really All About Non-Consensual Porn. Retrieved from Forbes: https://www.forbes.com/sites/daveywinder/2019/10/08/forget-2020-election-fake-news-deepfake-videos-are-all-about-the-porn/#26a929963f99

 

 

Will Artificial intelligence replace our doctors?

2

October

2018

5/5 (1) There is a worldwide shortage of doctors. More than half of the world population doesn’t have of has bad access to healthcare. The waiting lines are very long in a lot of places. AI could offer a solution here, giving more people access to health advice of good quality.

Although artificial intelligence (AI) is still in the early stages of testing and adoption in the healthcare space, many say it will have a huge impact in this field. Some even say it will gradually come to replace doctors.

Babylon Health, a company based in the United Kingdom, is testing an AI medical chatbot in Rwanda. It works like this: a patient enters information into the chatbot. The chatbot then aggregates the data and suggests solutions for the patient. It recommends the patient to see a doctor or to get a prescription rather than diagnosing him/her, although Babylon claims it could. Babylon also launched a site with the same idea, making it possible for people around the world to fill in their symptoms and get possible diagnosis.

Even though some are sceptical about the accuracy of the new ‘doctor’, the chatbot even passed mock medical exams with a higher score compared to a human doctor. Furthermore, in questions it had seen before, it had 98% accuracy, so once a machine learns something, it never forgets.

Although a lot of benefits are scientifically proven, some senior doctors are sceptical of the claims robots will replace humans, stating the human aspect of health will remain too important and can never fully be replaced by a robot.

 

What are your thoughts about this topic? Would you want to be seen by artificial intelligence instead of a human doctor? Do you think it could be possible that a machine can completely replace a doctor and is it ethical to replace doctors by artificial intelligence?

 

 

 

Babylon Health (2018). Babylon Health. [online] Available at: https://www.babylonhealth.com/news [Accessed 29 Sep. 2018].

Norman, A.(2018). Your future doctor may not be human. This is the rise of AI in medicine. [online] Futurism. Available at: https://futurism.com/ai-medicine-doctor [Accessed 30 Sep. 2018].

Vallancien, G.(2016). Tomorrow’s doctors will be replaced by machines, so their role will be that of advisor. [online] L’Atelier BNP Paribas. Available at: https://atelier.bnpparibas/en/health/article/tomorrow-s-doctors-replaced-machines-role-advisor [Accessed 29 Sep. 2018].

Wilson, C.(2018). Is an AI chatbot really better than a human doctor? [online] New Scientist. Available at: https://www.newscientist.com/article/2173056-is-an-ai-chatbot-really-better-than-a-human-doctor/ [Accessed 29 Sep. 2018].

Please rate this

The World’s First Clean Meatball

20

September

2018

No ratings yet. The agricultural industry, and more specifically the meat industry, is expected to change significantly in the upcoming decades (Le Mouël et al., 2015). Today’s society is facing numerous challenges regarding feeding the growing human population. Alexandratos and Bruinsma (2012) argue that a sixty percent increase in the world food supply is needed before 2050 to feed the unprecedented increasing human population. Besides the necessity of feeding the population, the agricultural industry must respond to developing social expectations as well as reducing the carbon footprint significantly (Hocquette, 2016). Society is becoming increasingly concerned with the life and welfare of animals, the threatened environment as well as a balanced and healthy diet. Further, as the agricultural industry is responsible for one-third of all human-caused greenhouse emissions reducing the industries carbon footprint is essential in limiting the effects of climate change (Gilbert, 2012). Summarizing, the agricultural and meat industry are unsustainable in its current form (Feenstra, 2013). The aforementioned factors will drive fast-paced changes in the industry, one of which is artificial meat.

Artificial meat, or lab-grown meat, is supposedly heading to your dinner table. Lab-grown meat is made by taking a muscle sample from the animal, after which stem cells are collected. The stem cells are dramatically multiplied which ‘allows them to differentiate into primitive fibers that then bulk up to form muscle tissue’ (Scientific American, 2018). Over the last decade several start-ups have attempted to turn artificial meat into an economically viable solution ranging from beef, poultry, pork to seafood. For example, Memphis Meat has attracted 17 million dollars in funding from multiple investors including Bill Gates and developed the world’s first clean meatball. Unfortunately, a quarter-pound of Memphis Meat’s ground beef costs $600 (which still is a significant reduction from $300.000 in 2013).

Economically viable lab-grown meat has multiple benefits. Not only does it generate employment through the creation of a new industry, it can help support the increasing demand for meat around the world. As stated in the first paragraph, the agricultural industry in its current form is not able to (carbon) effectively feed the growing world population. Furthermore, artificial meat would stimulate small-scale farming while decreasing factory farming. Factory farming is notorious for its poor conditions and lack of regard for the life and welfare of animals. In other words, artificial meat is superior from an ethical stand point as well. A point often overlooked are the health benefits associated with lab-grown meat. Besides being able to control the ratios of protein or fat in the meats, there is almost no need for antibiotics as the meat will be produced in a sterile environment. The US Food and Drug Administration argued that antibiotics in the diets of our livestock contribute significantly to the development of antibiotic-resistant bacteria (FDA, 2018). Lastly and most importantly, are the suitability benefits of lab-grown meats. Jacobsen (2017) states that making “1,000 kg of cultured meat takes 7 to 45% less energy, 78% to 96% lower greenhouse emissions, about 90% less water, and 99% less land.” Therefore, switching from factory farming to the laboratories could have a long-term positive impact on the environment.

The potential benefits of lab-grown meat are significant and widely shared. However, issues such as taste, safety, healthfulness, technological developments and costs are often underexposed. Although the artificial meat industry still has a long way to go and a lot of hurdles to cross, I’m willing to bet you a (lab-grown) burger that McDonald’s will sell clean meat in the upcoming decade.

Arthur Fortanier

Discussion
Would you eat artificial meat? Why (not)?

Sources
http:/www.scientificamerican.com/article/lab-grown-meat/
http:/www.reddit.com/r/explainlikeimfive/comments/7yg3ar/eli5_synthetic_meat/
http:/www.fda.gov/ForConsumers/ConsumerUpdates/ucm092810.htm
http:/www.futuresplatform.com/blog/are-we-ready-artificial-meat
http:/www.bestfoodfacts.org/is-synthetic-meat-in-our-future/
http:/iopscience.iop.org/article/10.1088/1748-9326/10/8/085010/meta
http:/www.sciencedirect.com/science/article/pii/S0306919213001012

Please rate this

Digital Transformation Project – Keadyn – Team 66

23

October

2016

No ratings yet. Disruption of the Venture Capital Industry

For the purpose of this assignment, Keadyn, a venture capital firm from Rotterdam was analyzed. For privacy reasons, the technological solution cannot be fully revealed, but the way Keadyn is disrupting the Dutch VC market is interesting on its own.

Keadyn operates in the technology ventures industry, with a focus on early stage and seed-stage startups. The firm aims to invest a total of up to €50 million in 3 years’ time. The company operates primarily the Netherlands, but sourcing across Europe and EFTA. Expected timeframe per investment is 3 to 6 years.

Traditional VC model

Before we delve into how the industry is disrupted, we can look into how traditional VCs work. Investors, also called Limited Partners (LPs), invest in a fund that is operated by a General Partner. Investors put money in the fund but typically cannot decide what will happen with their money. The decision belongs to the board of that particular VC. Thus, this type of investing is suitable for investors that do not want to be involved in the investment process and only pick up the profits at the maturity. It is worth mentioning that traditional VC’s return on investment often underperforms the market.

The commitment from investors in a venture capital firm is usually 10 years. The General Partner commits only 1%-2% to the fund, but is eligible to around 20% of the profits. On top of that, VC’s charge yearly management fees to the LPs of around 2% of the investments by those LPs.

New approach to investing

Keadyn’s business model differs from the traditional Venture Capital (VC) model. Keadyn’s model is more similar to “angel investing”. The partners of Keadyn currently invest in startups with their own capital. In the future they are planning to do investments in which key keep at least 20% of the stake they buy to themselves and offer the rest to external co-investors. Hence co-investors will be able to decide whether they want to join a particular investment, deal by deal. Keadyn charges investors a 6% deal fee and then 25% on the upside when the investment is liquidated. For example, when the initial stake of a co-investor in the investment was €20,000 and the investment is sold for €100,000, Keadyn will charge a carry fee of 25% from €80,000.

The value proposition of Keadyn is that it works “With the money, not for the money”. Since Keadyn’s partners commit funds to a startup upfront, there are no agency problems. Investors can then decide for themselves, whether or not they think the deal is a good one.

The role of Keadyn does not end with the investment. It is in Keadyn’s interest to advise the startups, so they have better chance of success in the market. For this, Keadyn utilizes its network of professionals in different fields, such as legal and tax law, marketing, team performance, and financial analysis.

Industry development

Keadyn is not the only VC fund that changes their approach to investing. Next to traditional VCs, there are new competitors with an interesting business model. Two competitors in particular are worth mentioning:

AngelList

AngelList is a platform where investors and startups meet. Investors that are part of this network can find a deal to invest in, which makes him a “lead” in that investment. Then, the investor can find other leads to help him to perform the due diligence, but he also has to find followers that will join in on the deal. AngelList itself helps with administrative procedures. For that, AngelList takes a fixed charge of 8,000 dollars for completing the deal and administration. Additionally, AngelList takes 5% carry of the investment profit, with the investment lead taking 15%.

Gust

Gust is a SaaS platform that connects investors (VC firms and Angel groups) with startups. It offers a CRM tool both for Investors and startups to help monitor the deal flow. Investors are able to track their portfolio, collaborate with other investors and track their investment related calendar. Startups are able to upload their business plan and track their pitches.

The future of VC industry

The VC industry is more and more affected by technological shifts. There is a trend of VC firms starting to make use of data science to speed up the deal making process as well as bring consistency into their decision-making.

Some examples of this new approach include Google Ventures and Correlation Ventures. While the former uses various data-driven algorithms to help them make investment decisions, the latter has built the “world’s largest, most comprehensive database of U.S. venture capital financings”, which includes key information about most of the pursued investments made over the past two decades and allows them to make predictive models as a guide for their own investments decisions in less than two weeks’ time.

These examples illustrate new trends that are becoming more popular within the venture capital industry, however, for the moment only big companies with huge investment funds can actually benefit from this data-driven approach.

 

By Adrianna Henc 463511, Eva Novotna 370931, Stephan Verhoeve 459523, Joanna Małkowicz 462688

Please rate this