Author Archives: Danylo Okseniuk

A serious step after the rise of AI: The World’s Largest GPU DePIN build on Solana

Reading Time: 2 minutes

With their unmatched computational power, Graphics Processing Units (GPUs) have become the unsung heroes of the artificial intelligence (AI) scene, driving developments. In this paper, we explore the revolutionary potential of GPUs, especially with reference to the largest GPU DePIN build on Solana system in the world. This innovative infrastructure is a prime example of how powerful GPUs can be in fostering AI innovation, completely changing the field of computational research and application. Come along as we explore the incredible relationship that exists between GPUs and AI, as well as how this massive system has the potential to push the frontiers of what is possible in terms of technology.

io.net is the project bringing AI to the world, with the mission to put together one million GPUs in a DePIN. In fact, it means that io.net has 73,000+ independent node operators around the world. The number of nodes is growing every day, so IO DePIN Network is a “Game Changer” in its sphere. However, such a big amount of users is caused by several advantages on the market, not only because of good marketing. Firstly, io.net users have unlimited flexibility to choose from the world’s best GPUs and customize. Secondly, users can access the io.net network in a matter of seconds. Thirdly, with high cost efficiency, io.net can be up to 90% cheaper than the competition. With a lot of experience, io.net is moving forward together with its community, despite high costs that were significant for them.

AI compute requirements have been growing tenfold every 18 months. OpenAI’s rental of over 300K CPUs and 10K GPUs for GPT-3 training marks just the beginning of this compute-intensive era. io.net was born out of need. Urgently, they envisioned and launched the DePIN revolution. The future, where computing power has no bounds, is welcomed by io.net. I am really interested in this project and believe that their motto: “Your need, our focus” exactly describes their idea.

Sources: https://io.net/about-us
https://www.youtube.com/watch?app=desktop&v=11zOFwaXkRA&ab_channel=DeFiDad
https://x.com/ionet_official?s=20






Tagged

Volvo – safe, reliable, innovative. AI in new safety technology of Volvo

Reading Time: 2 minutes

Everyone who know what is Volvo have no doubts that it is the safest car that you can find on the road. The reason why this car manufacturer has such a good reputation and market demand is hidden in the long history of the company. In fact, the most impressive thing in the successful brand history is the installing three-point safety seat belt into the serious production PV544. Nowadays every modern car has seat belts and only thanks to this million of people saved their lives. The Volvo’s inventions did not end there, and every decade their cars became safer.

Now, the company has not changed its attitude to the production of cars, but on the contrary now combines production with the use of artificial intelligence. Moreover, not only wants to produce cars that will be safe for passengers , but looks to safeguard the actual environment.

For instance, the new release of Volvo car’s fully electric flagship SUV- the Volvo EX90 featuring OnePilot: AI-powered software that will raise the safety of cars to a higher level.

And what this software will do?

Zenseact (that introduces OnePilot) main goal is to turn cars into perfect drivers – is based on how human will act in different situations. Different kinds of principles and set of methods that improve road safety, create the technology training the software to avoid unpleasant situations on the road, as people would do. All the cars with this technology will record real-life accidents and will be a step towards studying situations on the road. As a result, such software, which will be gradually updated, will make the Volvo EX90 even safer.

Volvo EX90 (2023) – oficjalne zdjęcia i informacje

In addition, there is also new lidar-based technology in the new Volvo, which makes it possible to drive the car in darkness. It can be able to detect different objects at a distance of up to 250 meters ahead. Artificial intelligence will do everything for you to avoid a dangerous situation.

To sum up, AI has again proven its usefulness. I will continue to follow the development of this technology. I would like to believe that all car manufacturers will use the same technologies as in Volvo EX90 in the coming years. Maybe after that we will not see so many terrible incidents on the road.

Sources:

https://www.gunthervolvocars.net/volvo-technology.htm

https://www.volvocars.com/intl/v/car-safety/safety-heritage

Artificial Intelligence is already in business

Reading Time: 2 minutes

First of all, AI it is the future of our life. Nowadays we can easily see how this field is important and what role it plays in world economy system.

Especially, this area is very attractive for venture capitalists. In 2022 they have ploughed $67 billion into firms that claim to specialise in AI, according to Pitchbook, a data firm. Starting from the middle of 2021, the share of deals worldwide involving AI-related startups increased by 17%. I must say that this is very big breakthrough for such a period of time. Therefore it is not surprising that between January and October, 28 new unicorns(private startups valued $1 billion or more) have been minted.

It is a huge competition between the companies which are desperate to get their hands on AI talents. Derek Zanutto of CapitalG ,notes that large had spend years collecting data and investing in related technologies. So now they want to use this huge amount of data and AI gives them different ways to do that.

Unsurprisingly, that all huge organisations use AI to improve their software. For example, today Google uses AI  to improve search results ,finish you sentences in Gmail and work out ways to cut the use of energy in its data centres, among other things.

Big companies quickly generate the plan how to sale some of Ai capabilities to their clients. Revenues from machine learning cloud service have doubled. In addition, upstart providers have wide spread, like Avidbots that leveraging data from a variety robot sensors.

In October Microsoft launched a tool which automatically wrangles data for users following prompts. All other huge companies may try to do something similar and several startups are already doing this. For example Google, presents in their video their first foundation model, which uses prompts to crunch numbers in spreadsheet and perform searches on property websites.

Other amazing thing that AI can do it is artificial colouring. In 2021 Nike bought a firm which uses such algorithms to create new sneakers design.

And the last example of how artificial intelligence is useful it is new technology of John Deere tractors which have some AI capabilities. This tractors can solve food problem in the word. IT is so important.

However, it is hard to say that AI is so profitable.  

According to the McKinsey Institute’s survey: quarter of respondents to the survey said that ai had benefited the bottom line (defined as a 5% boost to earnings). The share of firms seeing a large benefit (an increase in earnings by over 20%) is in the low single digits—and many of those are tech firms, says Michael Chui, who worked on the study.

To sum up, this sphere is developing every day and become more and more necessary, but now it is not so great for large organisations in terms of increasing profits.

Write your comments, what do you think about it.

Resources:

https://www.automate.org/tech-papers/avidbots-ai-platform

https://www.economist.com/business/2022/12/06/artificial-intelligence-is-permeating-business-at-last

Tagged

Crypto Infection for $62 billion. Spectacular collapse of FTX. Will they be crypto’s undoing?

Reading Time: 2 minutes

In 2017 cryptocurrency exchange Binance created the first of a new kind of blockchain-minted digital asset: BNB coin designed to reward customers for their activity on their platform. For example, trading and referring friends. It was a perfect model for building a successful and effective cryptocurrency business.

“A model for building a scalable, and impactful cryptocurrency business,” heralded CoinDesk editor Pete Rizzo in 2019, after Binance moved the coin to its own proprietary blockchain. “Unbelievable brilliance.”

Today, the only amazing thing about exchange tokens, like BNB, is that they inflated to $10 billion in value and have become the foundation upon of which the fast-growing digital-assets markets rest.

 The big failure in the crypto business of the former billionaire Sam Bankman-Fried’s was his FTX’s own exchange token, which traded under the symbol FT. According to Reuters Bankman-Fried had lent his trading company, Alameda, billions of dollars in FTX customer funds, collateralized by these FTT tokens. Despite the fact that this tokens were essentially invented as a way to offer trading discounts and other perks.

Reuters:

Bankman-Fried showed several spreadsheets to the heads of the company’s regulatory and legal teams that revealed FTX had moved around $10 billion in client funds from FTX to Alameda, the two people said. The spreadsheets displayed how much money FTX loaned to Alameda and what it was used for, they said.

At the peak in 2021 FTT had a market value of $9.6 billion, but FTT does not represent any equity ownership in the FTX company. You can think of these exchange tokens as reward points you might get as a frequent customer of McDonald’s. They have a value, but it doesn’t mean that you can buy a car using this tokens.

However, this tokens trades in different crypto exchanges just as stocks do on the New York Stock Exchange. About a week ago FTT traded for $26  and had a market capitalization of $3.5 billion. Nevertheless, after Bankman-Fried’s rival Changpeng Zhao, Binance’s billionaire founder, post a twit to say that he was planning to sell over $500 million of FTT , it sparked the crypto equivalent of a bank run. Based on this, today FTT sells for near $2.7, , and given FTX’s recent bankruptcy filing, it is likely headed to zero.

Unlike Bitcoin ,FTT’s were created with pre-mining and destroys the whole idea of cruptocurrency!

 “Effectively, all of the FTT tokens were owned by a collection of people and entities,” said Bankman-Fried.

Moreover, In terms of governance, exchange tokens, like loyalty reward programs, are completely under the control of the entity that issues or redeems them, even if they profess to stick to pre-arranged schedules for issuance or burning.

Here’s the 16 crypto and DeFi exchanges below created exchange tokens to attract and retain customers.

  

It’s a huge system thanks to which exchange earn big money every day. I hate this system and don’t think that it is normal that they trade like stocks and have a value of over $60 billion, none represent ownership in their affiliated platforms.

Resources:

https://www.reuters.com/markets/currencies/exclusive-least-1-billion-client-funds-missing-failed-crypto-firm-ftx-sources-2022-11-12/

https://www.forbes.com/sites/ninabambysheva/2022/11/14/the-looming-62-billion-crypto-contagion/?sh=27e6e5c961c3

Tagged

China’s social credit system. What is it? How is it related to AI?

Reading Time: 2 minutes

First of all, ‘The China social credit system’ ( first announced in 2014)  is broad regulatory framework that checks whether corporations ,individuals and governmental entities across China is reliable or not.

China social credit system (also known as ‘China’s Ranking System’) refers to a diverse network of initiatives aimed to increase trust in Chinese society.

The main goal is to make it easier for people and business to make the  correct decision. The high credit score will be a good indicator that this person can be trusted.

The system began with focus on financial creditworthiness ,but after that include compliments and legal violations.

Structure of this social credit system is very simple. Starting with the 1,000 points, lowest score is 600 points and a maximum of 1300 points. On the one hand, you can gain your points doing some charity work and other work that helps society. On the other hand, you can lose your points, because of your behavior in society. For example , posting anti-government massages or cheating in online games.

In my opinion, it is not good that you can lose your points and get punished ,because of your attitude to the government. In fact, it does not give you a feeling of complete freedom.

However, if you do all that is required to get 1300 points you will have a lot of different rewards  that is very useful in everyday life.

So, you have an opportunity to choose to be a ‘good’ citizen or ‘bad’ citizen.

But how to control this process? The fundamental building block of social building system is data. China have a really huge data that is gathered by regional, central and municipal government bodies. ‘Big data’ algorithms are then used to process that data in a meaningful manner.

There are a lot of CV cameras throughout the territory of China that record everything that every citizen does.

In 2019 the State Council released  ‘Guiding Opinions on Accelerating the Construction of a Social Credit System and Building a New Credit-based Supervisory Mechanism’. After that there was a need artificial intelligence to provide early warning of risky actors in need of extra regulatory attention.

It means that AI is ready to do such actions and will improve this big system.

To sum up, it is very interesting idea to create such system that has caused more than just public shaming. In addition, this emphasizes that China is a very  developed country that knows how to use AI and data to control the situation in the country.

Thank you for reading

References:

https://nhglobalpartners.com/china-social-credit-system-explained/

https://www.businessinsider.com/china-social-credit-system-punishments-and-rewards-explained-2018-4?IR=T

Flying cars by Alef-attractive project for forward-thinking investors

Reading Time: 2 minutes

In the Fall of 2015, 4 founders set a difficult goal to create a  car that will be able to move in the air. This task sounds unrealistic, but you should not underestimate these four absolute geniuses.

In fact, in 2016 they created their first prototype that was demonstrated to very popular and big-time businessmen, Tim Draper. After that he became the investor of this potentially successful company.

The next big step was to make these inventions not just a museum exhibit, but a real vehicle that can easily fly. Thus, they worked a lot and in 2018 this first full-size ‘skeleton’ ascended to the sky for the first time. Shortly after that, the car was able to autonomous flight, that is wonderful for the safe continuation of testing. Apart from the above, the ‘product’(first semi-public demo of real full-size flying car) was presented to the investors in very thoughtful way.

However, the main question arises: Will this project become popular? Obviously so, and it happened in 2022,when the company officially change their name to ‘Alef’. Alef’s Model A – 100 % electric flying car  with its own design and a seven-year history of developing and carefully improving every detail of the project attracted more and more investors. For example, Tim Draper, an early investor of Elon Musk and SpaceX, that backed long-range company with $3 million in seed money.

“I put more (money in) when I saw that they had created a small drone prototype that did exactly what they told me it would do,” he said in an email. “The design is extraordinary. The sides of the car become the wings when the plane goes horizontal.”

In an interview, Dukhovny, who is Alef’s CEO said that the plan is to sell Model A for $300,000,with it start in 2025.

“The whole car is the wing,” said Dukhovny.

In fact, this car will have range 200 miles(322km) and flying range of 100 miles.

What is more interesting Dukhovny has an even brilliant plan  for 2030: A proposed Model Z sedan, with a flight range of 200 miles and a driving range of 400 miles – and with a prize of $35,000

“This is not more complicated than a Toyota Corolla,” he said. “Our goal is to make sure it has the same price point.”

Sourses:

https://www.reuters.com/business/autos-transportation/flying-car-by-california-startup-alef-attracts-early-tesla-investor-2022-10-20/

https://alef.aero/story.html

Tagged