Kozminski Techblog

Reading Time: 2 minutes

In the times of the digital transformation, users can find and read the news from many different sources. Beginning from the digital newspapers and magazines like CNN or BBC, through platforms like Twitter or Facebook, or by scrolling news aggregators like Google News or Apple News. But now, a new player is entering the news market, namely, Artifact.

Artifact which name merges articles, facts and artificial intelligence was introduced by the co-founders of Instagram, Kevin Systrom and Mike Krieger, however the application is an individual venture, as they had both departed Facebook in 2018. Artifact is a personalized news provider which uses machine learning to understand your preferences and interests and will enable you to discuss particular articles with your friends. It was even named a TikTok for text, because “The app opens to a feed of popular articles chosen from a curated list of publishers ranging from leading news organizations like The New York Times to small-scale blogs about niche topics. Tap on articles that interest you, and Artifact will serve you similar posts and stories in the future, just as watching videos on TikTok’s For You page tunes its algorithm over time.” as The Verge[1] described the app. The user will be able to publish articles in the feed page and discuss them with friends in public or via private chat. Artifact is based on a different model in comparison with Twitter and Facebook as in their feed the content of people we follow is being shown, and Artifact will provide the staff by only using machine learning regardless of who followed and who your friends are.

Source: The Verge

While the app’s premise of presenting news in a visually appealing way is intriguing, there are some critical concerns about its functioning.

First, there is the issue of credibility. With the current amount of misinformation and fake news spreading rapidly online, it’s crucial for a news app to have a strong vetting process for the articles it showcases. While Artifact claims to use advanced algorithms, we will be able to see in the future, how effective this process really is in ensuring the accuracy and reliability of the presented news.

Secondly, there is the question of impartiality. With the app being created by two individuals with a background in social media, there is the risk that their personal biases and ideologies may influence the types of news being presented. Moreover, it raises concerns about the access to the data, and in this instance, it can make significant influence in the political environment.

Thirdly, by using machine learning and creating very personalized feed it can put users in the information bubbles what can lead to creation of extreme orientations and even stronger polarizations of societies.

Last but not least, I would like to mention the possible business model. Artifact plans to enable the advertising function obviously but also considers revenue sharing deals with publishers. It’s a very interesting aspect of providing the news from various platforms and publishers as it may raise concerns about the copyright rights and licenses.

Please let me know, what do you think about this app, and maybe you have any ideas and solutions for the possible problems.

---

[1] https://www.theverge.com/2023/1/31/23579552/artifact-instagram-cofounders-kevin-systrom-mike-krieger-news-app

Reading Time: 3 minutes

The sharing economy refers to the idea of sharing resources or services, typically through a platform or network, as an alternative to buying or owning them. This can include sharing things like cars, homes, tools, skills, and more. The sharing economy is often thought of as a way to make more efficient use of resources and to reduce waste, as well as a way to save money and generate income. Some examples of sharing economy are Airbnb or Uber.

The idea of sharing economy has started on a TEDxSydney conference when Rachel Bostman asked the audience “How many of you owns a power drill?”, and obviously almost everybody raised their hand. Later she continued, “That power drill will be used around 12 to 15 minutes in its entire lifetime. It’s kind of ridiculous, isn’t it? Because what you need is the hole, not the drill”[1] and then the solution of sharing economy appeared.

https://cordis.europa.eu/article/id/411546-understanding-the-role-of-governance-in-the-sharing-economy

The sharing economy seems to be a remarkable idea for saving costs, increasing the efficiency or access to resources and bringing environmental benefits. By sharing resources, individuals can save money on the costs of purchasing and maintaining goods. Moreover, sharing economy platforms can help match supply and demand more efficiently, leading to less waste and better utilization of resources and can make it easier for people to access goods and services that they might not otherwise be able to afford or that are not available in their area which also leads to community building.

However, the reality doesn’t seem as wonderful as we may thought, because instead of a collegial favor of giving someone a ride, which should be the basic assumption of sharing economy, the ride is provided by a professional Uber driver for a fee. The same situation is with platforms like Airbnb and many other companies which have been dominated by professionals such as Vinted, where the amount of clothes provided by small companies which buy them in second hands for extremely small amounts and later sell on Vinted is increasing. To face this problem, Vinted has come up with a regulation, that people who sell commercially will be blocked[2], however there might be concerns about checking it, since for instance how would they prove whether somebody bought those clothes in second hand or tries to sell his own old clothes to give them second life.

And here appears another problem concerning law and regulations, because many sharing economy platforms operate in a regulatory gray area, which can lead to issues with safety, quality of service and fair pricing. An example of pricing issue has appeared in Poland, where driving companies have begun to cull taxis by offering significantly lower prices and driving without licenses. Fortunately, the issue with licenses have been fixed, but is shows that usually the correct regulations are being created long after the appearance of the problem.

Another problem which arises, is the underdeveloped of rating and reputation system as there is no specified differences between grades. In my opinion, very often when we provide an opinion about an Uber ride, we don’t really distinguish the difference between 3,4 or 5 stars.

To sum up, in my opinion the idea of sharing economy has evolved into a new business model instead of promoting the collegial favor, and companies like Uber were using it only at the very beginning and later used it to promote themselves. However the basic idea seems to provide many benefits and opportunities for users, nevertheless probably will remain in small communities and groups due to the idea of commercializing it. Let me know what do you think about it in the section comments!

---

[1] https://www.fastcompany.com/3050775/the-sharing-economy-is-dead-and-we-killed-it

[2] https://wyborcza.biz/biznes/7,177151,27813463,zarabiaja-na-vinted-500-zl-dziennie-firma-zapowiada-blokowanie.html

Reading Time: 3 minutes

In recent years the market of voice assistants had significantly grown as some leading tech companies were striving to developing their voice assistants’ technologies. The market is being led my three technological giants: Google, Amazon and Apple, having respectively 81,5mln, 77,6mln and 71,6mln users[1]. Although those technologies are developing and gaining users, all of those companies are struggling to monetize their assistants. In this post I would like to focus and analyze the case of Amazons’ Alexa.

Probably most of you know what Alexa is, however if you do not, let me quickly explain. Alexa is an intelligent voice assistant which is connected to a cloud-based service with voice commands. According to Wikipedia[2], “the technology is largely based on a Polish speech synthesizer named Ivona, bought by Amazon in 2013”.  Customers may use it through Amazons’ Echo device, a hands-free speaker. In order to use the device, you need to “wake it up” simply by saying “Alexa”. Moreover, the device is known for the ability to amuse customers with short stories and having a great sense of humor.

According to Financial Times[3], Amazon is planning some huge layoffs, and Alexa unit is probably an area which will get hit hardest. Alexa was launched in 2014 and through all of those years was being strongly developed, what resulted in creation of one of the best voice assistants in the world. Alexa is able to perform more than 100000 skills globally, with the biggest skillset of 77000 in US[4]. The functionality ranges from voice interaction, making to-do lists, playing music and providing weather and news to controlling several smart devices, however the list is much longer.

As we may look at the data from previous years, Alexa had the biggest impact and a remarkable advantage on the voice assistant industry according to Voicebot[5] in 2019 and 2020. Moreover, in 2021 the shipments of Alexa reached 21.9mln of devices[6] and Amazon had a 44% market share (according to the same research, in 2020 the shipments reached 45mln devices and a 64% market share, however due to pandemic it is being treated as an outlier).

All of those data might have looked optimistic, however in 2021 sales dropped to 4mln which means 78.9% decrease year to year[7] and currently Amazon is planning layoffs. The biggest problem is that, through all these years they were struggling to monetize it, but unfortunately failed as the devices are being sold at cost[8]. The idea was to sell people a device which they can use not only to check weather and news but also to purchase products from Amazon. It seems to be a good idea, as the purchase was supposed to be done by voice, albeit people are wary of buying things without seeing them. The next idea was to give a possibility of ordering food or drinks from restaurants, unfortunately it also failed. In 2021 only 17.5% of consumers used voice shipping, which definitely isn’t enough to satisfy the creators of voice assistants.

As a voice assistant, Alexa has to listen all of the time, and that is what concerns over 33% of consumers, as Voicebot mentions. “Amazon’s Alexa collects more of your data than any other smart assistant”[9], to be precise 37 out of 48 possible parameters, and e.g., Google collects 28. That information also isn’t encouraging for consumers, specifically for those who are vulnerable about their privacy.

Other companies are also struggling with creating a profitable business model for their devices and I think that the nearest future will determine the existence of voice assistants. Let me know your thoughts about those devices, and whether you have any ideas to monetize them!

---

[1] https://dailyweb.pl/amazon-alexa-miala-uczynic-windows-inteligentniejszym-nie-wyszlo/

[2] https://en.wikipedia.org/wiki/Amazon_Alexa

[3] https://www.ft.com/content/19cb5d06-f60e-41a3-aab6-0fcc060e2604

[4] https://policyadvice.net/insurance/insights/amazon-alexa-statistics/

[5] https://voicebot.ai/2020/05/11/voice-industry-professionals-say-amazon-alexa-is-having-the-biggest-impact-followed-by-google-with-everyone-else-far-behind-new-report/

[6] https://www.pcmag.com/news/amazon-powered-smart-speakers-see-big-2021-shipment-decline

[7] https://www.pcmag.com/news/amazon-powered-smart-speakers-see-big-2021-shipment-decline

[8] https://arstechnica.com/gadgets/2022/11/amazon-alexa-is-a-colossal-failure-on-pace-to-lose-10-billion-this-year/

[9] https://www.pcmag.com/news/amazons-alexa-collects-more-of-your-data-than-any-other-smart-assistant

Reading Time: 2 minutes

Have you ever fallen asleep or felt very tired while driving a car? If no, you are probably a responsible driver. However, if yes, that is exactly what drowsy driving is. It’s also known as driver fatigue or tired driving. The most common cause of that is microsleep which “refers to periods of sleep that last from a few to several seconds. People who experience these episodes may doze off without realizing it. Some may have an episode in the middle of performing an important task”^[1].

According to research done by AAA Foundation^[2], “Drowsy driving accounts for about 100,000 crashes annually on the roadway, 71,000 injuries and 1,55o fatalities each year”. Moreover, in 2019 more than 1,200 drivers who were involved in fatal crashes reported being drowsy^[3]. All those statistics prove that drowsy driving may create some serious threats on the roads.

However, there might be a solution for that. Namely, a polish high school student, Maksymilian Paczyński had founded a system called FATIK (Fatigue Detection Model to Prevent Accidents). The programme analyses eyes closing, yawning and head stoop of the driver. Later those symptoms are being analysed by neural networks of the programme which in case of an emergency turns on the alarm to prevent from microsleep. It also enables to collect data in order to create a profile of the driver and works whether it is daytime or night-time driving.^[4]

Currently on the market we may find some other solutions for drowsy driving, for instance Mercedes Benz’s Attention Assistant; “ATTENTION ASSIST relies on an algorithm and specialized sensor to detect driver fatigue. During the first few minutes of your drive, the system analyzes your personal driving habits using over 70 different types of parameters. A key component of the sensor features the ability to record steering movements and steering speed.”^[5]. However, according to research done about drowsiness^[6], the physiological detection (which includes the eye movement and head swaying) occurred to have the highest accuracy rate, but this technology has not been fully developed yet.

The FATIK system is going to receive financing, so in the nearest future we will be able to track the development of this system. Let me know in the comments what do you think about this solution!

---

[1] https://www.healthline.com/health/microsleep

[2] https://www.bankrate.com/insurance/car/drowsy-driving-statistics/#drowsy-driving-statistics

[3] https://www.bankrate.com/insurance/car/drowsy-driving-statistics/#drowsy-driving-statistics

[4] https://www.thefirstnews.com/article/teen-scoops-top-award-for-groundbreaking-ai-programme-that-stops-drivers-dozing-off-at-the-wheel-26412

[5] https://www.mercedesbenzofeaston.com/mercedes-benz-attention-assist/

[6] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3571819/