Author Archives: Sardor Sheraliev

Next gen Telecom – built-in AI translator

Reading Time: 3 minutes

We’ve built a tool enabling everyone to understand and speak without any language barriers within just one app; Petr Malyukov, CEO of YOUS.

If you speak one language, and only, you dislike the fact that you can’t talk with ease to others, who speak in different languages, wether it is audio and video chats, even simply during phone calls or messages, you constantly have to rely on some intermediary translation tool. Today, living in the globalized and digitalized world, we might still have problems with speaking to others and understanding them even online, and why it should be an issue? Now individuals worldwide can easily switch between chatting, audio or video calls in any language pair, be it Italian – German or English – Hindi, thanks to YOUS. It works just as easily as it sounds: if we are talking voice translation, the tool recognizes speech, interprets it, and speaks in a synthesized voice in the interlocutor’s language. Additionally, the whole conversation is duplicated in text chat format.

Another issue with interpreters is privacy and safety.
When translation services are hired in the corporate world, data confidentiality is granted by legal tools, like NDAs. It doesn’t work in personal lives, reason being, many of us not comfortable with speaking on private and intimate matters with a third party involved. Altogether with the AI-powered concept, YOUS allows users not only to eliminate disclosing sensitive information to simultaneous interpreters, but also to avoid any storage of their personal data.

YOUS quite practical tool to solve an ever – present issue that somehow hadn’t been tackled yet, and quite surprising that product is based on a body of present technologies merged in a hybrid system. YOUS can help to resolve an issue like primary communication apps, but with a completely new quality of breaking down language barriers. You can see in it all is clearly a massive shift in the communications landscape. We all are used to assorted machine translation tools and features that really do help us understand each other no matter the language, but there is immense room to improve, quicken, and sharpen what we have now. YOUS web app is just the beginning, and the capacity ahead is boundless. YOUS also has an app that enables calling mobile and landline numbers with a built-in AI translator, and at the moment YOUS working on developing the know – how for automated recognition of gender and language by voice, as well as voice cloning for cross – language synthesis.

Envision a world where individuals, enterprises, institutions, and other communication parties can finally chat one-on-one independently of their language.

Thousands of doors would get opened immediately. It is exactly the world we’re heading to when introducing a new AI-powered communication tool to the market. Whether you ever wished to understand and be understood immediately while chatting with your foreign friends or business partners online, or whether you never thought of this idea, the new era had arrived. YOUS holds the idea that, the time has come to change international communications by the adoption of AI into the everyday processes of our digital lives.

Resources:
YOUS
ProductHunt

Autonomous transport future is not arriving soon

Reading Time: 3 minutes

“Autonomy Day”, April 2019, the event hosted by Tesla. Tesla felt that everything has been so focused on the Model 3 and the rest of the company’s story has been lost. Elon Musk was confident about bringing forward fully – autonomous robotaxis by the end of 2020. However, we can see that the prediction was too optimistic as self – driving vehicles are still quite challenging to achieve.

What is a Robotaxi anyways?
We can define it as a fully – autonomous vehicle that can pick up and drop passengers at their destinations without any need of a human driver. Also, we can accept that drivers are the reason for the high costs of transportation services at some level. It is believed that autonomous cars can help bring down these costs and make transportation more affordable for every use case. However, unlike games and apps, launching an autonomous taxi has real risks that can cause injury or even death. Software – defined architecture and Nvidia’s chips optimized for deep neural networks are the heart of autonomous vehicles. On the other hand, it is not easy to just deploy this infrastructure and get it going hurled. Launching a robotaxi is still an idea for the far off future.

At the same event, Pete Bannon, the designer of the full self – driving chip, described the system, what timeline was for the system, and design requirements. The biggest takeaway, Tesla did look at the problem and work to solve specifically the issues they had for chips, which you can do only by designing a new chip, which is something most companies don’t have the luxury to do, as it is extremely expensive.

Tesla’s Robotaxi
It is became a real possibility talk after Musk claimed in 2019. Musk stated; Next year for sure, we will have over a million robotaxis on the road. The fleet wakes up with an over – the – air update. That’s all it takes.

It only created hype in the market. The reason is that Tesla already has thousands of models running on the road deployed with eight cameras each to collect video recordings from all angles. Industry experts still doubted any possibilities of such autonomous vehicles, as the Waymo’s CEO John Krafcik said; For us, Tesla is not a competitor at all. We manufacture a completely autonomous driving system. Tesla is an automaker that is developing a high quality driver assistance system.
He also pointed out, that a company can’t keep improving its driver assistance system and hope for it to become fully – autonomous one day. However, Musk responded to these comments saying; To my surprise, Tesla has better AI technology, hardware, and software than Waymo.
And Tesla’s robotaxi continues to be under the microscope of professionals thanks to the high expectations set by Elon. While he believes that Tesla is quite close to SAE Level 5 full autonomous, his promise keep backfiring at him.

What is the Market share of Robotaxi?
It is rising market, expected to be worth $1.03 billion in 2023 and rise to $38.61 billion in 2030. The pandemic showed a massive downfall in the transportation industry as drivers could not risk by taking random passengers. Customers also believe an infected driver is a high – risk carrier. The industry is highly reliant on drivers, and the need for robotaxis is real case. This triggered a rise of investors interested resulting in the increase of market value. The major shareholders of this market include:
Waymo, Lyft, Uber, Tasle itself, Ford and a few others.
The road safety and concerns for security control is the reason why autonomous vehicles are being prioritized. This gives robotaxis another market advantage as they are fueled by cameras and sensors, ensuring no risks are taken. These factors ensure the rise of share value for the robotaxi market.

Who is working on Robotaxis beside these giant players?
There are other competitors in the market working on the same venture for sure. Autonomous transportation service in the form of robotaxi has become a race between players in the industry. Hyundai, Cruise, Yandex, Didi, Baidu and others are some companies working on the idea. All companies are progressing quickly, with most of them heavily relying on the Nvidia chips. It is one side to develop an autonomous car capable of driving itself, however, it is totally different from being able to deploy it immediately. The complications involved in this matter are grave, and Tesla might be thinking way over its capabilities. Once they are ready, they most likely face regulatory challenges from authorities that will not allow an act of such high risk. Moreover, the are severe challenges in relevance to insurance regularities as who takes blame in the event of an accident.

“We are ten years out before we see fleets of robotaxis. They ain’t coming anytime soon, even with the most advanced sensing systems in the world”, said the Lucid’s CEO, Rawilson.

Resources:
Robotaxi fleets
Market size
SAE Level 5
Autonomy Day

Unlocking The Machine Economy

Reading Time: 3 minutes

MachineFi, is the financialization of machines, or, more precisely: internet-connected devices. So far, we only knew about “DeFi” Decentralized Finance and “GameFi” Game Finance, even Helium has a decentralized market for device connectivity.

However, why do devices need financialization or decentralized markets? When Amazon’s internal cash register rings in the sale of internet-connected devices like Alexas or Apple Watches, it marks the beginning—not the end—of the financial contract between buyer and seller. This is why tech companies are willing to sell devices at, or even below, cost. In other words, the financialization of internet-connected devices already exists in a rudimentary way.

So MachineFi takes the riches generated by device-financialization from big-tech and decentralizes it to actual users, which is what web 3.0 is all about: changing the financial model of the internet (and now devices) from platform advertising to user ownership (often through the broad distribution of a token). Example: At a congressional hearing on crytpocurrency, “Bitfury” CEO Brian Brooks, explains the blockchain and how it plays an integral role in Web 3.0.

Now let’s dive into how MachineFi works, by analogy, with the U.S. mortgage market.

The value generated by the “internet of things” from now until 2030, according to McKinsey, will be $12.6 trillion. And while this isn’t a one-to-one comparison, the primary US mortgage market was just under $17 trillion as of early 2021. Which gives a sense of the opportunity. In other words: Machines could become as important to household wealth as homeownership. This is a radical claim. But one that seems plausible given the rate of automation. The problem is, without MachineFi, this wealth will accrue to the balance sheets of a few powerful tech giants. The average person risks being shunted outside the economic arena in the face of automation. MachineFi is about addressing this inequality—giving everyone the financial tools to participate in our economic future.

MachineFi and U.S. Mortgages

The modern mortgage was introduced in the United States in the early 1930s when less than half of Americans owned a home. After the introduction of the mortgage, the rate jumped to 65 percent and has stayed roughly steady ever since. As the rate of ownership rose, the value of the average home rose with it (in 2000 dollars) from $30,600 in 1940 to $119,600 in 2000. The point here is that financial innovation was a cause not an effect of this boon to household wealth. But not all financial innovation is created equal (not all financial innovation results in equitable distributions, obviously). With 2007-2008 financial crisis, Collateralized Debt Obligations (CDOs) and Mortgage-Backed Securities spun out of control as mortgages were chopped up and tranched into piles of financial abstraction dynamite that blew once housing prices started to fall. In this analogy, MachineFi is to the mortgage what big tech is to the CDO. MachineFi, which is done right, will cause the proliferation of device innovation and distribution of wealth to households. This is why it needs to be taken under consideration, right before (and during) the economy engages in this seismic shift to automation.

Why MachineFi Matters?
MachineFi’s main objective is to transition traditional IoT and machine verticals into MachineFi decentralized applications (Dapps) that will enable millions of users to participate in the machine economy with billions of smart devices. The world has witnessed the boom of smart devices and automated machines, including smart home devices and smart cars. However, few have noticed inter-machine communication is significantly increasing. According to a McKinsey report, the Internet of Things (IoT) could unlock a global economic value of up to $12.6 trillion by 2030. Techjury estimates that over 125 billion devices will be connected to the Internet by the start of the next decade, powering that machine economy and make world better place to live.

Resources:
CISION PR Newswire
YouTube

Amazon Web Services toward reducing Carbon Emissions

Reading Time: 3 minutes

As Amazon claims, they currently have more than 234 solar and wind energy projects throughout the world, including projects in Finland, France, Germany, Ireland, Italy, the Netherlands, Spain, Sweden, and the UK. Together, these projects will provide over 10,000 MW of global renewable energy capacity when complete.
They are inviting organizations to join The Climate Pledge, a commitment to becoming net-zero carbon by 2040, 10 years ahead of the Paris Agreement. Co-founded by Amazon, more than 200 organizations have signed so far The Climate Pledge, committing to use their scale to decarbonize the economy through real business change and innovation.

Great! How renewable energy further reduces carbon emissions?
451 Research, a global research firm found that if a 1-megawatt corporate data center (about 1,000 square meters, at an assumed 30% electrical utilization rate) switches their applications to the cloud, they could reduce emissions by about 1,079 metric tons of carbon dioxide per year. This is equivalent to removing over 500 cars from the roads or offsetting the annual electricity emissions of over 50 average households across Europe. That number rises to as much as 1,293 metric tons of carbon dioxide when a cloud provider is powered by 100% renewable energy. AWS is on path to achieve 100% renewable energy by 2025.

451 Research also find out and AWS commissioned that, migrating compute workloads to AWS across Europe could decrease greenhouse gas emissions equal to the footprint of millions of households. In addition, businesses could potentially reduce carbon emissions of an average workload by up to 96% when AWS reaches its goal of purchasing 100% of its energy from renewable sources.
And how? They surveyed senior stakeholders at over 300 companies across a broad range of industries in France, Germany, Ireland, Spain, and Sweden operating their own data centers. Results show that companies in Europe can reduce their energy use by almost 80% by moving their compute workloads out of on-premises data centers to AWS. Companies could potentially further reduce carbon emissions from an average workload – by up to 96% – once AWS meets its goal to be powered by 100% renewable energy. And compared to the computing resources of the average European company, cloud servers are roughly three times more energy efficient, and AWS data centers are up to five times more energy efficient. In fact, moving a megawatt (MW) of a typical compute workload from a European organization’s data center to AWS Cloud could reduce carbon emissions by up to 1,079 metric tons of carbon dioxide per year.
Kelly Morgan, Research Director, Datacenter Infrastructure & Services at 451 Research, part of S&P Global Market Intelligence said; We were struck by how much opportunity there is for European businesses to increase energy efficiency and reduce emissions by looking at their IT infrastructure. If you think of the electricity consumed and emissions produced by tens of thousands of companies across Europe operating their own data centers, this is an area that appears to be overlooked.

We can conclude that, cloud data centers are efficient because cloud providers focus on sustainability across their entire operation. In contrast, most businesses don’t prioritize their data center infrastructure sustainability, which means their data centers are not optimized for efficiency. Most businesses don’t prioritize addressing data center energy costs and carbon emissions because digital infrastructure is not their core business. Cloud providers focus on efficiency as a best practice. They operate servers at much higher utilization rates and design facilities to use less energy and water. Overall, the AWS Cloud infrastructure is up to five times more energy efficient than typical EU enterprise infrastructure. AWS achieves lower energy use in its data centers through designing cooling systems that reduce energy and water use, and using real-time sensor data to adapt to changing weather conditions. AWS’s scale enables high resource usage, and its global cloud infrastructure is built using Amazon’s own custom hardware, purpose-built and optimized for workloads run by AWS customers. This includes power-efficient processors like Graviton 2, the ARM-based AWS – designed chip that provides better performance per watt than any other Elastic Compute processor.

How it can affect and change the current situation?
AWS has responsibilities to the communities where they operate, and to them, that means sustainability and environmental stewardship. AWS is continuously working on ways to increase the energy efficiency of facilities and equipment, as well as innovating the design and manufacture of servers, storage, and networking equipment to reduce resource use and limit waste.

Resource: 451 Research report

Next – Generation Technologies aren’t public yet

Reading Time: 6 minutes

Have you ever catch yourself regularly mentioning technologies like Quantum, Graphene, Digital Twins and most recently Meta (Facebook) over the dinner and among friends and have become the go – to person for questions about future innovations? These technologies no longer hold any secret for us and we heard, seen them a lot.

Technology, is ever – changing, and this precious knowledge must be both managed and updated all the time. With this in mind, let’s explore and put together with you friends, the order of technologies that are likely to make big waves and huge impact on our lives, however are not on the public network radar as of today.

#1. Zero – knowledge proof

Computer scientists are perfecting a “cryptographic tool” we could use to prove something without revealing the information underlying the proof. It sounds incredible, however, not impossible once you wrap your head around the concept and the fact that it is a bit more complex than saying “Come on brother, you know I am good for it”.

Let’s simplify through an example.

Imagine, a man, John Doe has a blind friend Jane Doe. He also has in his possession two marbles of different colors, though they are identical in shape and size. Jane puts them behind her back and shows one to John. She then does it again, either changing the marble or showing the same thing once again, asking if this is the same as the marble previously shown. If John were guessing whether it was the same or not, he would have a 50/50 chance of getting it right, so she does it again, many times. Because John sees the marbles’ colors, he gets it right each time, and the chance that the guessed lucky diminishes. Jane thus knows that John knows which marble is the original shown without her ever knowing the color of any of the marbles. Boom, zero – knowledge proof! Obviously, it all gets mathematic and cryptographic from here, however you get the gist.

How it can affect and change the world?

It is easy to come up with cool use cases, for instance, if an application needs to know that you have enough money to put a transaction through: your bank could communicate that yes, that is the case, without giving an amount. It could also help identify a person without a birth certificate, allow someone to enter a restricted web resource without needing to display their date of birth or help with nuclear disarmament. Additionally, it could provide proof of a crypto – currency transaction without revealing its amount, as BitCoin does, and always will be.

Resource: Wikipedia

#2. 20 – minute Water

Not all technology needs to be high; there is often something rather elegant in low – technology solutions to complex issues. 20 – minute water is one of these solutions, promising to provide clean drinking water to the masses. All one needs to do, is soak a piece of cotton (which is highly conductive) for 20 minutes in an inexpensive solution containing carbon nanotubes and silver nanowires, then connect to it two electric wires to pass a little current through it. Twenty volts is enough to instantly electrocute bacteria and make the water passing through this filter drinkable, without the need for the electricity – thirsty pumps that are used throughout the developing world. The silver then nullifies anything that is not been killed by the electrical current.

How it can affect and change the world?

We can save remote areas where people do not have access to chemical treatments such as chlorine and they are mostly developing countries. It could save some of the 300 000 children under 5 who die worldwide every year because of waterborne diseases such as the cholera, typhoid, and hepatitis. It could also help some of the 2.2 billion people who do not have a wastewater treatment system.

Resource: Stanford News

#3. Robotic bees

It is not great out there, climate – wise. And that is killing bees, which we need to pollinate 35% of our crops, which we also need for food, for not being dead. Are we going to stop climate change to save the bees, I guess we can’t and that is not how we roll. Instead, we are possible going to create robotic bees to pollinate plants just as the real things do.

Find out how Walmart did it. Details are scarce, however, most researchers estimate, that the bees would work by attaching horse hair coated with ionic liquid gel to a tiny drone. The hair, picks – up pollen from one flower, and moves it to next one. Researchers at Harvard have long been working on “RoboBees” using such techniques. What Walmart offers on top, is a wide array of sensors, cameras, artificial intelligence to locate the relevant crops and pollinate them as needed.

How it can affect and change the world?

Costs. Operating such a technology will make us see autonomous insect pollinate large fields in the coming years, which could save thousands of farmers from ruin, and ensure we can still have almond milk on the shelves.

Resource: The Wall Street Journal

#4. Two – Photon Lithography

3D printing is still a solution looking for a problem, having failed to find its target audience over the past decade. On the one hand, 3D printers are still too expensive for the average John Doe; on the other hand, they are not sophisticated and fast enough for large – scale manufacturing companies. This may change over the next few years: researchers have developed a method that uses lasers to project millions of points simultaneously onto 3D – printing material, instead of using one point at a time. And because they are bad at branding, they called it “Femtosecond Projection TPL”. To easily understand FP-TPL, simply imagine using a million heated needles to strategically melt a block of wax versus using a single needle, which means that incredibly tiny structures can be 3D printed much faster, while still ensuring a good quality of build.

How it can affect and change the world?

Teams working on the innovation seem focused on flexible electronics and micro – optics. However, quick discoveries around materials (both liquid and solid) have led researchers to think over that we can be able to build small but imagination – baffling structures in the near future. Once the quality can be controlled over large scales, one could easily imagine this technology being used for the creation of healthcare – related nano robots, allowing for the treatment of a multitude of diseases on the molecular level.

Resource: Science Journal

#5. YOLO version 5

Real – time object detection is a technique used to detect objects from video / clip. It is the underlying technology behind… well, most things we want to use in the future, from Tesla’s self – driving cars to Amazon’s cashless stores. The YOLO “You Only Look Once” models refer to some of the most versatile and famous real – time object detection and labelling models.

The latest iteration “version 5” of the model is worthy for the couple of reasons. It is written in PyTorch, which will make its deployment to mobile a lot easier. It is also fast, pretty quick. 140 frames per second, while also preserving accuracy. Finally, YOLOv5 is nearly 90 percent smaller than YOLOv4.

How it can affect and change the world?

This innovation is important in terms of to deal with real – time analysis; detecting and labelling video files 140 times per second. For context, previous models struggled to get to 10 frames per second. At this speed, we can use AI technologies on video of fields as varied as the world of medicine or that of sport. It can also improve detecting obstacles in autonomous cars earlier to avoid even more collisions. Big deal, huh?

Resource: Github repository

What do I disagree with and what I can counterbalance?

  1. We have to solve the issue with zero – knowledge proof imperfection. The minimization of destroying or changing the messages while sending to verifier or prover by building knowledge around and listening to feedback from our community.
  2. We have mentioned the 20 – minute water technology and the amount of silver used for the the nanowires is so small the cost in negligible, and the electricity needed can be easily supplied by a small solar panel or a couple of 12 – volt car batteries. Since the filter does not trap bacteria (killing the instead), it can have much larger pores, allowing water to speed through at a more rapid rate. More than 80 000 times faster than existing filters, to be exact. And it does so without clogging, an issue which plagues existing solutions.
  3. I have a few gripes with Robo bees. First thing is, small budget farmers might never be able to afford robotic bees, and we would once again be empowering the BIG GUY against the little one. We also do not know how these bees would impact the fauna overall, both on and off the fields. How about if we try to save the real bees? Open – ended questions.