According to local media outlet Sina News, the platform was established to enable the trade of NFTs as a secondary market by the state-owned Chinese Technology Exchange, the state-owned Art Exhibitions China, and Huban Digital Copyrights Ltd, a private business organization. The Platform will also provide institutions and individuals services for copyright protection and rights monitoring with regard to digital assets.
Yu Jianing, an expert in digital assets and Chinese metaverse space, claims that the establishment of the China Digital Assets Exchange marks an acceleration of the country’s cultural industry’s digital transition. Jianing believes the “China Digital Asset Trading Platform” draws on the characteristics of a national-level exchange of the China Technology Exchange, performs trading duties on behalf of the exchange, and develops a thorough trading system, a standardized trading system, and a scientific trading system for the digital asset trading industry.
In recent years, the market for NFTs in China has expanded dramatically, with numerous high-profile transactions occurring on different platforms. However, the absence of a centralized market has hindered buyers and sellers from swiftly finding and exchanging NFTs. There have also been questions regarding the legitimacy and ownership of some NFTs.
In order to overcome these problems, since 2021, cryptocurrency exchanges have been prohibited in China, though crypto ownership is regarded as virtual property that is protected by the law. In the same year, the Chinese government declared it was creating a national NFT platform. The marketplace aims to offer a centralized platform for buying and selling NFTs as well as a means of effectively regulating the NFT market and safeguarding the rights of producers and users.
The Hangzhou Internet Court, which specializes in internet-related legal matters, ruled on November 29 that digital assets like NFTs are virtual properties that are legally protected and that they possess the qualities of property rights, such as value, scarcity, controllability, and tradeability. Prior to this, the Chinese central bank unveiled its intentions for the CBDC and made trial versions of mobile applications for digital yuan wallets available.
The Securities and Exchange Commission of Pakistan (SECP) on Wednesday prohibited digital lending platforms from sending borrowers’ data outside of Pakistan and limited their ability to use coercive means to recover debts.
The SECP has instructed the digital lenders through Circular No. 15 that the borrowers’ data cannot be stored on any cloud infrastructure that is under the control of a country other than Pakistan. The SECP commission has also published new guidelines for digital lending that Non-Banking Finance Companies (NBFCs) using digital channels or mobile applications must follow. The regulator made these announcements in response to growing concerns about mis-selling, data privacy violations, and intrusive recovery tactics used by licensed digital lending companies.
Before a loan is granted to the borrower, the standards specify the bare minimum obligatory disclosures and key fact statements (KFS), which must include the loan amount approved, annual percentage rates, loan term, installment/lump sum payment amounts with dates, and any fees and charges.
The non-banking financial companies would be expected to communicate these important details to customers via audio, video, emails, and text messages in both English and Urdu. This is important to ensure transparency and ease of understanding, says the regulator. It further stated that any fees not covered by KFS would not be passed along to the borrower.
The new SECP digital lending guidelines will require a licensed digital lender to publish on its lending platform(s) or app(s) its full corporate name and licensing status, ensuring that any advertisement and publication is fair and do not contain any misleading information.
The SECP has also detailed a thorough grievance redressal framework in addition to the current NBFC grievance redressal structure.
Additionally, even with the borrower’s consent, the digital lender will not be permitted access to the borrower’s phone book, contacts list, or photo gallery to protect the data’s confidentiality and privacy, the SECP informed. Other than those who have been specifically authorized by a borrower as guarantors and have also given their authorization to the digital lender at the time of loan acceptance, the rules have also prohibited the lenders from contacting the individuals in the borrowers’ contact lists.
Creta, a web3 video game developing company, reveals four novel blockchain-based games for its web3 ecosystem. The game is called “Kingdom Under Fire: The Rise.” It comprises a metaverse, a one-stop gaming destination, and a Super Club blockchain community service.
Creta is a renowned game developer, popular in the Korean market and supported by several industry leaders like the Yield Games Guild (Japan) and Yoshiki Okamoto (the company behind Monster Strike).
Creta announced the games at Creta Summit 2022, held in Tokyo, Japan, an event that several blockchain enthusiasts and web3 gamers attended. Creta showcased the gameplay footage of the new blockchain games to unravel the look and feel of the avatar-based setup.
Ray Nakazato, Chief Creative Officer of Creta, said that after a while, some characters and collectibles present in the game will be turned into non-fungible tokens (NFTs) and can be traded. He added, “The game will be playable in seasons, and data will reset at the end of each season with an opportunity to transfer player NFTs to a new season.”
Kingdom Under Fire and other games developed by Creta will be released and played via the company’s tailor made gaming platform similar to web2 games, but with an additional metaverse layer. This layer would allow social interaction and economic features for NFT utilization.
The CoinSwitch Web3 Discovery Fund has evaluated 150 Indian web3 startups across different use cases like infrastructure-related products and blockchain analytics within three months of opening up for applications. It is expected to announce its first cohort of 10 startups soon and plans to invest in more than 100 startups by the end of 2024.
In the early stage of the fund, CoinSwitch will be writing cheques of $100,000-$200,000 and working as a medium to introduce startups and funds jointly in larger rounds with its 20 Venture Capital partners. The partners participating in the funding round will be Ribbit Captial, Coinbase Ventures, Tiger Global, Woodstock Fund, Sequoia Captial, Elevation Capital, and incubation partner Builders Tribe.
Ashish Singhal, Co-founder and CEO of CoinSwitch mentioned that CoinSwitch has already evaluated 150 startups. CoinSwitch is unable to keep pace with the kind of talent and innovation they are seeing coming in. He also stated that it is hard for retail users and businesses to search the Web3 ecosystem, so people have started building the Web3 equivalent of Web2 companies such as Google, for crypto, and more.
Besides CoinSwitch, its companions like CoinDesk have rushed into venture capital with CoinDCX Ventures targeting early-stage Web3 startups. This might sound like a blessing for the Web3 ecosystem, but it also highlights the challenges in the pure-play crypto segment.
Parth Chaturvedi, the head of CoinSwitch’s Web3 Discovery fund, stated that the fund’s applications included many use cases around blockchain analytics, self-custodial solutions for Web3 wallets, Web3 infrastructure-related products, and more.
Tech giant Baidu which operates China’s largest search engine, announced it has expanded the commercial operation area and hours of its driverless taxi service in Wuhan, Central China. The deployment of robotaxis at night in Wuhan by the company’s autonomous ride-hailing platform Baidu Apollo Go heralds a new phase in the commercial use of autonomous driving in China.
The public will be able to use the robotaxis at Junshan New City in the Wuhan Economic and Technological Development Zone between 7 am and 11 pm starting this week, according to a statement from Baidu on Monday. Its autonomous vehicles could previously only be used in the city from 9 am to 5 pm. The revised program is anticipated to serve one million customers in selected neighborhoods of Wuhan, a metropolis of more than ten million people.
One of the major technological challenges with autonomous driving has always been the nighttime environment since it is difficult for cars to discern objects and pedestrians in low light. After beginning to use a combination of external cameras, radars, and lidars to improve visibility in poor visibility conditions, the company made the latest announcement.
Apollo Go now operates over 50 completely autonomous taxis in Wuhan, covering an area of more than 130 square kilometers. Last month, Baidu published a concept for its sixth-generation electric robotaxi, the Apollo RT6 EV, a hybrid between an SUV and a minivan with a removable steering wheel.
In the past two decades, there has been a lot of interest in autonomous driving because to its numerous advantages, like relieving drivers from exhausting driving and reducing traffic congestion, among others. As a result, researchers have paid close attention to autonomous vehicles due to their potential to increase the efficiency and safety of transportation networks through control algorithms while cutting down on fuel usage.
Despite encouraging advancement, ramp merging has been a major challenge that threatens to cause frequent traffic jams on the road, higher fuel consumption and emissions, safety concerns, and rear-end and side collisions. This is due to the decision-making process of merging cars, which causes them to first slow down or even stop on the ramp before merging into the main lane at an appropriate moment through control without interfering with the moving vehicles on the main lane. Since the cut-in movements of ramp vehicles can frequently disrupt the mainline traffic flow and result in numerous issues, ramp merging is crucial for freeway traffic operation.
At present, with their real-time communication and precise motion control abilities, autonomous vehicles can improve ramp merging activities through enhanced coordination techniques. Using specialized short-range radio communications and cellular networks, the communication technologies enable detailed and rapid information transmission among road users, traffic infrastructures, and control centers. As a result, vehicular moves can be arranged through real-time interactions among traffic participants. Furthermore, because they are less prone to delays and mistakes in the processes of identification, decision-making, and performance, autonomous driving systems in cars can execute the intended actions in a steady and timely way.
To enhance tactical decision-making in autonomous vehicles, a number of impediments still need to be addressed, and as computational resources advance, there will undoubtedly be a number of exciting new chances to solve challenging issues. In an effort to boost efficiency, researchers from Carnegie Mellon University have created a reinforcement learning (RL)-based framework that could aid in the performance of autonomous vehicles in ramp merging settings. Their framework, outlined in a pre-published paper on arXiv, can contribute to strengthening the safety of autonomous vehicles at these crucial decision-making periods while lowering the likelihood of accidents.
Reinforcement learning is one of the most important machine learning methods to achieve Artificial General Intelligence (AGI). RL systems are frequently trained in gaming environments, which serve as testbeds for teaching agents new tasks using visual signals and the popular “carrot and stick” approach.
In a reinforcement learning approach, artificial intelligence (AI) agents are put into simulated settings and given two options that are determined by predetermined policy. The agent makes a decision and is “punished” or “rewarded” for it; in other words, positive actions are encouraged, and negative ones are discouraged. Whether its decision has a favorable or detrimental consequence, the AI modifies its policy accordingly and repeats the process with fresh choices made that are time influenced by the modified policy. The AI agents keep going through this process until they find the best solution.
Given the potential outcomes of the infinite complexity of complicated real-world circumstances and the significant risks involved, RL may require fundamental technological advancements to enable complete ‘autonomous’ driving. In recent times, reinforcement learning has been extensively investigated for lane-changing decision-making in AVs, with positive results. However, it was eventually discovered that most of these studies had compromised either the safety or the efficiency of the algorithm.
According to Soumith Udatha, one of the researchers who created the model, Prof. John Dolan’s department at CMU has been working on numerous autonomous driving applications for quite some time. Udatha says that due to the challenges posed by fast-moving cars, drivers with different driving styles, and inherent uncertainties, the application on which his team concentrated in this work is freeway merging.
The central goal of Udatha and his team’s study is to increase the safety of autonomous vehicles. In their paper, they sought to develop a framework particularly designed to capture ramp merging situations and plan a vehicle’s actions based on its analysis of any uncertainties and potential dangers.
Though, as mentioned above: reinforcement learning models interact with the environment and gather information to maximize their rewards, Udatha explained that this data exploration meets with several complications when used in practical contexts. This is partially due to the fact that not all of the states the agent encounters are safe. In order to assure safety at a given distance, the team limited its RL policy using control barrier functions (CBFs). As a result, they disregard unsafe states and improve a system’s capacity to learn how to travel according to environmental constraints.
CBFs are a group of relatively recent computational techniques created to improve the reliable control of autonomous systems, by ensuring a suitably-defined barrier function remains bounded for all time. They can be leveraged directly for a variety of optimization issues, particularly ramp-merging. Although they look good on paper, the optimizations they carry out do not take into consideration the information a system gathers as it is exploring an area. Reinforcement learning methods, as per Udatha, can eliminate this discrepancy.
The research team discovered that their algorithm could be applied to RL settings that are both online (while interacting with the environment) and offline (learning from a fixed dataset or logged data). However, offline reinforcement learning has currently become a core approach for using RL methods in practical settings. This is because it allows for the empirical evaluation of RL algorithms based on how well they can use a predefined dataset of interactions and produce real-world effects.
The team used a dataset extracted from the NGSIM Database, which includes high-quality traffic data at four locations: two freeway segments (I-80 and US-101) and two arterial segments (Lankershim Boulevard and Peachtree Street), between 2005 and 2006. The datasets collected and created for each location comprise the vehicle trajectory data (primary data), various location-specific primary and support data (e.g., ortho-rectified pictures of the research area, Computer-Aided Design (CAD) drawings of the study area, signal timings, weather data, detector data), raw video files, and processed video files.
Because the team now has massive volumes of data for offline RL, training on offline datasets may eventually result in superior models. The researchers also found — using their metrics — that adding probabilistic CBFs as limitations improves safety by partially accounting for driver uncertainty.
Using the online CARLA simulator created by a group of researchers at Intel Labs and the Computer Vision Center in Barcelona, Udatha and his colleagues put their framework through a number of tests. Their approach produced outstanding results in these simulations, emphasizing its great implications for boosting the safety of autonomous cars during ramp merging.
The research team now intends to continue the study by training their model to merge an autonomous car with many other vehicles in a situation with unknown drivers. Additionally, they discovered that there is presently no benchmark that can be used to evaluate different ramp-merging strategies, therefore Udatha’s team is working on creating one for NGSIM.
IIT Madras’s Bachelor of Science courses (BS) in data science and applications won the silver prize in the ‘Best Online Program’ category at the Wharton-QS Reimagine Education Awards. At the same time, NPTEL (National Programme on Technology Enhanced Learning), a joint initiative of IITs and IISC, won the gold prize in the ‘Lifelong Learning Category’ at the Wharton-QS Reimagine Education Awards.
The Wharton-QS Reimagine education award is called the ‘Oscars of Education.’ It recognizes and celebrates the outstanding achievements of educators, institutions, and organizations that drive innovation and excellence in education. The award ceremony was held on December 7th and 8th at the Wharton Campus, Philadelphia, USA.
IIT Madras BS program has more than 15,000 students enrolled currently. The program is in a hybrid mode that has online delivery and in-person assessments. At the same time, NPTEL offers more than 4000 courses for certification currently. It has more than two crore enrollments and about 23 lakh examination registrations.
Professor V. Kamakoti, Director of IIT Madras, mentioned that the BS program and NPTEL are examples of how technology can deliver high-quality education to students from different locations. IITs are committed to innovating and finding new ways to enhance their students’ learning experience.
Meta has agreed to pay $725 million for settling a class-action lawsuit that claims the company inappropriately shared users’ information with a data analytics firm used by the Trump campaign called Cambridge Analytica.
In 2018, it came to light that the information of up to 87 million people might have been inappropriately accessed by the third-party firm, which had filed for bankruptcy in 2018.
According to the plaintiffs’ lawyers in a court filing, this is the most significant recovery ever in a data privacy class-action lawsuit and the highest amount Facebook has paid to settle a private class-action.
Meta has not admitted wrongdoing and claims its users consented to the practices, thus suffering no actual damages. Meta spokesperson Dina El-Kassaby Luce said that the settlement was for the good of its community and shareholders. The company has since then revamped its approach to privacy, she added.
According to the Plaintiffs’ lawyers, about 250 million to 280 million people might be eligible for payments for the class action settlement. The amount of individual payments depends on the number of people who come forward with claims that are valid.
Over the past several years, Facebook’s data leak to Cambridge Analytica has sparked global backlash and government investigations into the company’s privacy practices.
According to a California Highway Patrol (CHP) traffic crash report, a Tesla driver claimed that their vehicle’s “Full Self-Driving (FSD)” software unexpectedly braked and caused an eight-car pileup in the Yerba Buena Tunnel last month. Nine people were treated for minor injuries, including one child who was hospitalized. The CHP analyzed tunnel footage and discovered that the Tesla performed an illegal lane change before suddenly slowing down from 55 mph to 20 mph, forcing vehicles behind it to collide with one another, according to the December 7th report received by CNN Business.
The above incident can be viewed as another incident caused by the automaker’s US$15,000 Full Self-Driving software package. But can we still afford to ignore questioning the facade of hype behind the technology that is being marketed to bring the next revolution in the autonomous vehicle industry? Is it possible that beyond the promise to advance the scope of self-driving vehicles, Tesla might have overstated the capabilities of its full-self driving software?
Tesla currently comes with a standard driving assistance feature dubbed Autopilot in all of its new vehicles. Additionally, it offers extra functions, viz., Smart Summon, Navigate on Autopilot, and Automatic Lane Changes, in a package that is commonly marketed as Full Self-Driving. Under its FSD Beta program, the company also permits select owners to access and test features — which have not yet been entirely bug-fixed — on public roads. The software is designed to keep up with traffic, navigate within the lane and adhere to traffic signals. It requires a careful human driver who is ready to take over complete control of the vehicle at any time. While some drivers have been thrilled by this software, many are concerned that a Tesla outfitted with FSD would misbehave at the worst possible time.
The “full self-driving” beta, which became available to everyone in North America since November, has proven to be worrisome for many Tesla customers who paid US$15,000 for the software upgrade, believing in Tesla CEO Musk Elon’s claims. This is because the program occasionally tries to strike curbs or travel on the wrong side of the road. While Tesla is continually improving the technology and addressing its flaws, beta testers’ experiences offer a glimpse into the incredibly risky and expensive gamble the company is placing on its so-called full self-driving technology.
The ‘full-self driving’ was initially conceptualized as a technology that can help vehicles maneuver through the roads without any human assistance. However, if Tesla requires human drivers to take over in case of any malfunction of its “FSD software,” the term has already lost its original meaning. In order to rekindle the hype around fully autonomous vehicles, major companies like Tesla, Waymo announced that ‘full-self driving’ would be gradually introduced, starting with testing the vehicles in geographically restricted areas. Though this year, we have witnessed many companies testing autonomous robo-taxis in streets of Las Vegas, Phoenix, and Los Angeles, it will take decades for them to expand to a level that even comes close to countrywide deployment at the current rate. Consequently, costs and the organizational learning curve have been substantially higher and longer than anticipated. The companies have discovered that the technological prerequisites to allow mass adoption of the technology are far more challenging than they had initially anticipated, despite investing billions of dollars in its development over the course of a decade.
In hindsight, Tesla’s “Full Self-Driving” software is more akin to a “Level 2” advanced driver assistance system that requires constant active supervision by a driver, despite the advertising. There is concrete evidence to back this claim too! The DMV and Tesla were exchanging emails in 2019 and 2020 that were disclosed by Plainsite in response to a public records request revealing the company’s Full Self-Driving mode, also known as City Streets, was a Level 2 technology while Musk was making audacious claims about fully autonomous vehicles. This proves that Tesla’s technology is no more capable of autonomous driving than rival driver-assistance systems offered by companies under the Level 2 category.
If you account for the reality of the situation, why does Tesla continues to market itself as the developer of ‘full-self driving’ software for its vehicles? Is it blatant ignorance to steer the sales by capitalizing on the misinformed hype or an optimistic bet? While Tesla doesn’t claim the software to enable fully autonomous driving, does it set a dangerous precedent?
For now, we have a temporary yet effective solution to this dilemma: California lawmakers have recently passed a new law prohibiting Tesla from labeling its software ‘Full Self-Driving!’ The new law, sponsored by Democratic state Sen. Lena Gonzalez of Long Beach and signed by Gov. Gavin Newsom this legislative session, prevents car dealers and manufacturers in California from “deceptively naming or marketing” a car as self-driving if it’s outfitted with only partial automation features which still necessitate human drivers to pay attention and handle driving.
Gonzalez informed the Los Angeles Times that the state Department of Motor Vehicles already has regulations against the misleading advertising of self-driving vehicles. However, the DMV’s lack of enforcement pushed her and state legislators to introduce legislation to incorporate the standards into state law.
The new bill, Senate Bill 1398, is one of the hundreds of new state regulations that will go into effect in 2023. It explicitly targets Tesla’s promotion of software contained in some Tesla models that implies the car can fully drive itself. According to Gonzalez, the bill increases consumer safety by mandating dealers and manufacturers who sell new passenger vehicles equipped with a semiautonomous driving assistance feature to include a comprehensive description of the capabilities and limitations of those systems.
It is important to note that the new bill does not address the safety concerns surrounding the Full-self driving software. However, it is the most recent instance of politicians, regulators, and customers fighting back against what they claim to be false and misleading advertising. In response, Tesla fought against the law, claiming that it already makes Tesla owners aware of the limits of the Full Self-Driving software.
As tensions mount, Tesla may have to come clean about their litany of bogus claims about rolling ‘Level 5 autonomous vehicles,’ made by Musk every year. The takeaway is simple: Tesla cars must be subject to the same testing regulations as other autonomous vehicles that are now on our roads if they are sufficiently automatic to be advertised as Full Self-Driving. If the cars are not sufficiently automated to be regulated as autonomous vehicles, Tesla should be barred from marketing the technology as Full Self-Driving. Therefore, the California government is right in asking companies like Tesla to refrain from misleading people under the pretense of offering fully autonomous technologies.
Though, banning Tesla from advertising vehicles as self-driving if they still require driver supervision is a historic milestone, much needs to be done. Even if a car is capable of operating safely in all circumstances, drivers will still need to be on guard and prepared to take over if necessary.
Tesla has chosen to make its self-driving technology available to consumers, unlike other self-driving car companies like Waymo and Cruise, who test their vehicles in carefully monitored pilot projects. To minimize the risk of regular drivers facing risks of accidents or software malfunction, NHTSA should come up with a preapproval system before installation. It should also come up with certifications, as DMV offers to run autonomous vehicles in California, before the four-wheelers hit the roads. These are important as the self-driving automobile currently lacks a real industry software and hardware standard.
While companies like Tesla are aiming for fully autonomous driving, it does not imply eliminating the scope of driver assistance. In addition to sending frequent information on crashes and instances to DMV, where the human driver had to take over to prevent a crash, Tesla cars must have a certified and trained test driver operating the vehicle. Further, NHTSA should come up with some regulations that allow it to take action whenever Tesla launches software updates or recalls software features, irrespective of mode – via the internet or directly to drivers. This can address the governance blindspot that arises when autonomous vehicle companies add new features or patch software flaws remotely, triggering concerns about liability, accountability, and safety.
While Google’s AI translation service works remarkably for many languages, people find it lacking in understanding Bengali nuances and expressions. Following the same thought, a non-profit organization called Bengali.AI works on making Bangla typing and translation inclusive with the help of researchers and AI professionals.
Bengali.AI recently started its largest project called Bangla Speech Recognition. The goal is to teach computers to understand Bangla. For this, they conducted a social media campaign called the “Bok Bok Campaign.” Bangla speakers from around the world contributed their voice data to this campaign to expand the Bengali.AI voice dataset.
The idea behind Bengali.AI saw the light in 2017 when a group of students from Bangladesh started the project with a vision to push AI research in Bangla translation. Founded by Ahmed Imtiaz Humayun and his peers, the platform became the destination for those who dream.
While it was a novel idea, the Bengali.AI platform did not receive much recognition until late 2019, when the organization tied up with Google to launch several competitions on Bangla Graphmemes (segments of word formation). Bengali.AI could pull over 7.5m hours of research work with this collaboration.
With developments in Bengali.AI over the next few years, the platform now works like Grammarly (an American typing assistant) but for the Bengali language.
