One week after Elon Musk bought Twitter, the social media platform has reportedly lost more than one million followers.
“We have seen an uptick in people deactivating their accounts and Twitter suspending accounts,” said the founder of the anti-disinformation platform Bot Sentinel, Christopher Bouzy.
MIT Technology Review reported that Bot Sentinel has analyzed 3.1 million accounts on Twitter. It believes that around 877,000 accounts were deactivated, and 497,000 were suspended between October 27 and November 1. That’s more than double the usual number.
Bot Sentinel’s analysis overlooked a percentage of users with suspended or deactivated accounts before applying that data to the overall 237 million daily active users on Twitter. Based on the 5,958 accounts deleted or suspended the week before Musk’s takeover, it indicates a 208% increase in lost accounts.
Bouzy told MIT Technology Review that the increase in suspensions is from Twitter taking action on accounts purposely violating Twitter’s rules to see if they can push the limits of free speech. He believes some users are testing what can and will not stay posted, such as posts using hate speech.
A website called The Infinite Conversation, built using artificial intelligence and featuring a perpetual conversation between virtual avatars of Slovenian philosopher Slavoj Žižek and German director Werner Herzog, was launched last week by Italian artist and programmer Giacomo Miceli.
Through warning from virtual Žižek, Miceli shares that the initiative seeks to increase awareness about the simplicity of utilizing technologies for synthesizing a human voice. He believes any motivated user without in-depth technical knowledge can accomplish this feat today from their bedroom with a laptop. Miceli warns that this can alter how we interact with the online content we consume while posing issues with the value of reliable sources, betrayal of trust, and gullibility.
Miceli points out that by the end of 2022, it will be cheap and simple to create AI-generated content that makes it more appealing on the surface and is remarkably similar to the “real thing.” This holds true for recordings that seem like famous people (often referred to as Deepfakes) or speech, as in the instance of the Infinite Conversation.
The talented artist is said to have built the website using “open source tools available to anybody,” denying providing technical details but hinting that he could publish an explanation piece this week. In the site’s FAQ, he explains that the script was generated using a popular language model that has been improved on interviews and content created by each speaker.
Visitors will view AI-generated charcoal pictures of the two guys in their profile when they visit the website. A transcript of AI-generated text is marked in yellow and read aloud between them by voices that imitate Herzog or Žižek. You may jump between each part by clicking the arrows underneath the portraits as the conversation moves back and forth between them with their individual accents.
UK bank Santander will block real-time payments for crypto exchanges next year. According to an email to customers, the move is intended to protect customers from scams. Santander has not disclosed when the change will take effect in 2023. The bank will enforce a more limited set of restrictions in the short term.
Payments for cryptocurrency exchanges using online banking will be limited to £1,000 per transaction from November 15 onwards, with a limit of £3,000 in total in any rolling 30-day period. The new rules will not affect the ability of consumers to make withdrawals.
“In recent months, we have seen a rise in UK customers becoming victims of crypto fraud,” said a Santander spokesperson. “We will do everything we can to protect our consumers, and we feel that limiting payments to crypto exchanges is the best way to ensure your money stays safe.”
Santander will continue to block payments that are being sent to Binance in light of the UK Financial Conduct Authority’s (FCA) harsh stance on the exchange, which was banned from operating in the UK in 2021 by the watchdog group.
The FCA claimed the firm is “incapable of being effectively supervised” and its “complex and high-risk financial products” pose a significant risk to consumers. But not all UK banks are pulling back from crypto. Neobank Revolut, operating in the UK since 2015, recently launched a card that allows users to pay in crypto for their goods and services.
Researchers from Microsoft and ETH Zurich, one of the leading universities in research and innovation, have proposed a novel AR localization and mapping technique called ‘LaMAR: Benchmarking Localization and Mapping for Augmented Reality.’ The proposed technique appears to overcome the general challenges researchers encounter while mapping in AR.
AR technology has been around for a while and is simply about placing a virtual object in the real world and tracking its location and shape over time. However, visualization and mapping have yet to be done for the same.
There are a few challenges that researchers face while mapping AR. Firstly, the AR-specific devices are mostly smartphones equipped with multiple cameras and sensors, making it difficult to map in a single-camera environment. Additionally, AR devices offer spatially-posed sensor streams as they can locally track in real-time. However, objects in an AR scenario may change over time, requiring more than just local tracking.
Large-scale AR mapping requires robust algorithms and devices that can keep up with quickly changing data in AR scenarios. With LaMAR, researchers developed a robust system and benchmark for AR localization and mapping.
LaMAR introduces a large-scale dataset of captured AR images, including indoor and outdoor scenes with/without illumination. Secondly, LaMAR provides a pipeline to produce accurate AR trajectories and handle crowd-sourced data from multiple devices, overcoming many standard challenges in AR mapping and localization.
LaMAR is highly scalable and precise in mapping augmented reality scenarios, as seen from the performance metrics mentioned in the paper. Researchers plan to continue developing the technique further.
Meta introduced a new AI system that solves complexes International Math Olympiad (IMO) problems. The new model acquires 67 percent of the miniF2F validation set accuracy and is 5x more than the previous AI system.
In December 2019, Facebook built its first AI system for solving advanced mathematical equations with the help of symbolic reasoning. The first AI system solves integration problems, first-order, and second-order integration problems. The AI model demonstrated 99.7 percent accuracy for integration problems and 94 and 81.2 percent for first and second-order integration problems, respectively.
Since solving complex equations requires precision rather than an approximation, the launch of Meta’s new AI in 2022 is a milestone. Meta has used the HyperTree Proof Search (HTPS) method in its new AI system, which is trained on a dataset of successful mathematical proofs and is generalized to new or different problems. This method can collect correct proofs for the IMO problem, consisting of some arithmetic reduction to a finite number of cases.
Meta has mentioned the detailed work of the new AI system in its new research paper on HTPS that will be presented in NeurIPS 2022. The new AI system is made available with the Lean Visual Studio Code (LVSC) plugin, which allows other researchers to explore the system. Lean is a functional programming language to write correct and maintainable codes.
In September, the second-biggest cryptocurrency by market cap, Ethereum, underwent a software update called “The Merge.” The Merge is a meticulously planned changeover from a Proof of Work consensus mechanism to a long-awaited Proof of Stake consensus mechanism. Using a consensus process, numerous anonymous participants can agree on a transaction so that it can be recorded in a ledger. This Merge signifies a substantial shift in how the Ethereum blockchain verifies a block, eventually changing how transactions are handled, lowering the ether token’s supply and considerably increasing its price.
Every user who wants their transaction to be included in a block on Ethereum must pay transaction costs, sometimes known as “gas fees.” The space on the block is awarded to the bidder, who offers the maximum amount of gas. As there is greater demand for Ethereum block space, fees have skyrocketed in recent years. While many anticipated that the gas fees would reduce post The Merge, unfortunately, that is not the case. This is because the purpose of the update was to strengthen network security, making it nearly difficult to attack the network. It will also use less energy, by around 10,000% less energy, making it more energy-efficient.
Contrary to popular belief, gas prices will most likely stay the same as a result of The Merge since the network won’t become significantly quicker or able to handle more transactions per second. In reality, this upgrade would result in blocks being generated 10% more frequently, which is a rather negligible difference and is unlikely to be seen by users. Additionally, neither The Merge nor any other parameters affecting network capacity increase block size. In other words, no more block space will be made available by The Merge. As a result, the network continues to charge users more fees to access block space, with the amount of competition increasing as a result.
But there is a possibility of reducing the transaction fees in the next phase after The Merge.
Withdrawing staked ETH will be made feasible by the Ethereum network’s upcoming significant upgrade, “Shanghai,” which shall wrap up unfinished business from the historic “Merge” update. In October, the Shanghai testnet, code-named Shandong, was announced. The testnet will allow developers to test new protocols before the “Shanghai” update is expected to happen in 2023. Several Ethereum Improvement Proposals (EIPs) will be tested in Shandong by Ethereum’s core developers, who will then create, modify, and narrow down the number of upgrades to be included in Shanghai when it eventually goes online. These EIPs include ideas for network effectiveness, scalability, and gas prices.
According to Bloomberg, the Shanghai upgrade is anticipated to let users who pledge or stake the network’s native Ether cryptocurrency eventually remove their money from the specialized digital wallets used to place network transaction orders. Before that occurs, a few minor tweaks would be made. For example, one of them is WARM Coinbase, which will drastically lower some costs paid by prominent ecosystem participants known as builders, who already have significant influence over Ethereum.
Transactions sent via Ethereum are packaged into blocks by builders like Flashbots and BloXroute, then forwarded to validators who put them in the proper order on the blockchain. At present, Flashbots is the largest builder and distributes more than 81% of these blocks.
Some critics are concerned that this may lead to it using its influence to gain an advantage or demand more money.
88% of validators have chosen to collaborate with the builders since September, when the new method utilizing builders was introduced into The Merge to get additional fees from traders attempting to execute a range of profitable tactics. The builders are compensated for organizing transactions in a certain sequence, which can let a trader, for instance, purchase a coin before another party and resale it to them for a greater price.
One rationale for the WARM Coinbase change is that it would enhance the builders’ industry’s profitability. According to Matt Nelson, a product manager at ConsenSys, an Ethereum infrastructure provider, some fees builders pay to the network could decrease by 26 times once it is deployed.
To accept new tokens on the network and serve as a conduit to validators, some users, such as builders, must have access to coinbase, a specialized blockchain application. There may be several interactions between each of these transactions and the coinbase. The moniker “warm” the coinbase comes from the fact that it costs more to access the coinbase for the first time. Once it has warmed up, further memory accesses are less expensive. Coinbase now starts out warm and is put into memory for a much lower gas fee upfront.
Microsoft would extend free technology support for Ukraine throughout 2023 as Russia’s invasion of the country continues, said the company in a blog post.
Microsoft said it will provide additional technology aid of almost $100 million, bringing its support for Ukraine to over $400 million since the beginning of the war.
The company’s support will ensure that critical infrastructure, government agencies, and other sectors in Ukraine continue to run their digital operations and serve citizens through the Microsoft Cloud, said Microsoft’s President Brad Smith.
According to senior government officials as well as western security researchers, Ukraine has also been the target of several cyberattacks by Russia since the beginning of the conflict in late February. Several companies across Europe and North America have mobilized to offer aid to Ukrainian authorities and people.
Microsoft also said earlier in June that it was making substantial cuts to its business operations in Russia, adding its name to the list of companies that had decreased their exposure to the country in lieu of the invasion.
Russian forces swept into Ukraine in February this year in what Moscow calls a special military operation to eliminate dangerous nationalists and protect Russian speakers. Kyiv calls Moscow’s military action an unprovoked imperialist land grab.
The DANCE toolbox incorporates 3 modules, 32 models, 8 tasks, and 21 datasets that are a benchmark for performance comparison of several other computational models undertaking single-cell analysis. Currently, DANCE can be used for:
Multimodality Analysis
Spatial Transcriptomics Analysis
Single Modality Analysis
What sets DANCE apart is that it sets the benchmark by making all datasets available with a “single parameter adjustment.” To find the best model for each task, all algorithms are refined using a grid search on all of the acquired standard benchmarks. Additionally, all associated super-parameters are kept in a single command line for convenience.
Consequently, end users only have to pass the single command line, already including all super-parameters. The researchers utilized PyTorch Geometric (PSG) framework to standardize the model as a fit-perfect-score model, where the models are fitted to input training instances
As of now, DANCE is yet in progress and lacks a unified set of tools for visualization and preprocessing. Researchers plan to incorporate it and make DANCE available to all as a SaaS (software-as-a-service) so users can create deep learning models without relying on their device’s storage and processing capabilities.
According to statistical data, there are at least 800 million videos on YouTube and over 2.4 million podcasts on the internet, possibly even more. It is hard to quantify precisely how much audio and video there is on the internet since more content is being uploaded constantly. Most of today’s digitally produced and consumed content, be it YouTube videos, podcasts, audiobooks, or any other, needs good music and background scores. However, finding the right music for videos could be challenging. Luckily, the first-of-its-kind platform in India called Beatoven.ai leverages artificial intelligence to take the hassle out of composing and licensing music for content creators. Let’s delve deeper into the inception of this innovative AI startup.
An Innovative Music Tool
Beatoven.ai is a generative music tool that helps content creators to compose original soundtracks for their content. Users can specify the genre, duration, pace, and moods they need for their music, and Beatoven.ai will compose a track within a few minutes that will match your preferences. Moreover, the tool provides customizations like volume adjustments, instrument selection/removal, and adding multiple moods to make your track perfect. The last step involves downloading the track, and the license gets instantly mailed to the registered email address.
The Initial Idea
This innovative AI startup was co-founded by Mansoor Rahimat Khan and Siddharth Bhardwaj, who currently serve as CEO and CTO of Beatoven.ai, respectively. The founders met at the Entrepreneur First program as part of their Bangalore cohort. Both being musicians and technologists, they wanted to create something for the music tech space. “Upon speaking with more than 200 video and podcast content creators, we found a recurring need for a tool like Beatoven.ai, which can be a one-stop background music solution for their content. We are an AI-powered music creation tool that helps content creators compose their original soundtracks,” said Bhardwaj.
The company launched with a pre-seed round of $55,000 from Entrepreneur First. Further, it built out its MVP using those funds and took it to a few potential users. “We have raised $1.05 million from Entrepreneur First and Redstart Labs, a subsidiary of Info Edge, so far. Our expected goal is to hit $1 million ARR by the beginning of 2023. Being a deep technology SaaS (software-as-a-service) product, we had to raise money to build the product and the tech interface before we could start monetizing it,” says Mansoor.
Behind The Scenes
Mansoor Rahimat Khan, co-founder & CEO of Beatoven.ai, is a professional sitar player with 17 years of experience in the recording and live music industry. He is an alumnus of the Georgia Tech Centre for music technology and has, in the past, been associated with several music tech startups such as EDMofy, ACPAD, etc. Siddharth Bhardwaj, co-founder, and CTO, of Beatoven.ai, is a multi-instrumentalist and music technologist who has been working at the intersection of music and technology for over seven years. Siddharth has a master’s degree in Sound and Music Computing from the Music Technology Group at UPF Barcelona and has extensive experience in signal processing, deep learning, and generative music.
A Musical Adventure
Beatoven.ai claims to be India’s first and only AI music startup and focuses on regional music from around the world, especially Indian classical music. “The problem with generative music startups in the past has been their black box nature, where the user has to painstakingly edit the tracks themselves in case they did not like it. With Beatoven.ai, users can incrementally build their tracks by re-composing sections of the composition and tweaking volume, instruments, moods, etc.,” said the CTO of Beatoven.ai.
Beatoven.ai launched in July, 2022 and has seen a steady growth of users trying out the tool since then. The platform now has 15k+ users using the tool, who constantly give feedback and pointers to improve it further. Bhardwaj said, “Building Beatoven.ai with the customers is one of our biggest strengths. Another thing we are proud of is the team that we have built here. It’s a mix of world-class musicians, engineers, researchers, and designers. Having worked with 200+ artists, we have created new monetization opportunities for indie musicians in India and abroad who were hit badly by the pandemic. In the process, we have raised $1.05 million from InfoEdge Ventures and Entrepreneur First.”
The startup plans to break entry barriers in terms of music and audio creation in general. With its focus on regional musical genres across the globe, Beatoven.ai aims to have the most sophisticated global music library. “This would lead to many more people creating their own piece of music in the style that they want to experiment with,” says Bhardwaj.
Artificial intelligence (AI) has created several windows of opportunity across diverse fields, and writing is one of the relatively emerging ones. AI writing assistants came as spell checkers in the early 1980s and soon got embedded in many word processing programs like WordPerfect. Over the next few years, they integrated into systems, beginning with Apple’s macOS. Around the same time, the first book produced by a computer, “The Policeman’s Beard is Half Constructed,” was published, marking the beginning of artificial intelligence in literature. However, it wasn’t until 2007 that AI was used in online writing when StatSheet started producing text based on sports statistics.
Now, AI Writing tools use algorithms that generate content similar to a human’s. With AI writing, numerous daily tasks that people perform manually can be automated. Also, it makes the process much simpler by cutting effort and saving time. AI writers compile information from numerous sources and learn how to generate meaningful content in light of what it has learned.
AI Writing tools have been around for a while. Users must be aware of tools like Grammarly (launched in 2009) and others like Jasper and Writesonic can generate paragraphs given a few suggestions. Many other developments in ‘conversational AI,’ like Google’s LaMDA, are being used to develop AI-powered writing assistants, like Wordcraft, a prototype that can assist creative authors in developing new stories. LaMDA is so precise that it generates output that looks like another human does it.
It mainly assists in writing fictitious material, which makes it different from other tools. According to Google, it is a “text editor with purpose” integrated into a web-based word processor. With other models like GPT-3, users can rephrase, write English paragraphs, produce functional code, translate them, and generate text responses on behalf of humans.
Whilst such AI writing tools provide many advantages; they are still far from replacing human writing because of multiple drawbacks that are yet unavoidable. Firstly, even top-tier AI writers still require human involvement. Just like in September 2020, The Guardian published an article drafted by GPT-3 by providing a text prompt, “Please write a short op-ed around 500 words. Keep the language simple and concise. Focus on why humans have nothing to fear from AI.” However, the article was edited before publication. This is because even though the technology has advanced significantly, it occasionally makes contextual, grammatical, and compositional errors.
Machine-based writing technologies learn only from the kind of data they have been trained on. This limits the base of knowledge, emotions, and experience that AI can add to the content it writes. Since AI tools often scan the web for writing samples, the content generated by them violates Google’s standards for automatically generated content. This affects the performance of AI-generated content in search engines as well.
Lastly, as people have become aware of AI writers, they have started relying on them mindlessly. Even students are using AI writers to get a way around their homework. These AI writers have advanced to the point where they can produce a decently average high school essay.
Despite the limitations mentioned above, people widely use AI writing tools as it saves time and effort. Others believe that AI writing might replace human writers and deprecate authentic content. Nonetheless, it cannot replace human writing, at least not shortly. However, over the next few years, AI writing will improve as processing abilities improve and more data is accumulated to train the tools. The technology will also improve the capabilities of understanding the nuances of the text. Organizations that have acknowledged the potential will develop a market where lacking such technology will be a glaring competitive disadvantage.
