If you are facing difficulties while working on Microsoft Excel, you now have an AI Bot that can create formulas and solve your problems. The bot aims to make working with Excel easier for those who are not familiar and do not really get the hang of Excel formulas.
Getting pro at all Microsoft Excel’s ubiquitous features is a daunting task. Given the massive amount of data you add, it is not a very intuitive spreadsheet software for an average person. But, the new AI Bot is there to help you.
Simply head to the website and summarize the problem you are willing to solve. The results are even more accurate if your prompt is as specific as possible. For instance, refer to exact cells, rows, or columns while giving inputs. The site explains that you might need to rephrase your problem if you do not get the desired results.
The website also mentions an example for users to understand how to give inputs. It says, “When column A equals the sum of ‘hello,’ calculate the sum of column B.” In Excel, it outputs to “=SUMIF(A:A, “hello”,B:B).”
The project is free and open to donations. It is currently accepting donations to fund future development of the work to generate more and more Excel formulas. So far, it has generated over 3,22,000 formulas, and the count is increasing.
Apple’s AI team is bringing a new AI system, GAUDI. The AI system can generate 3D indoor scenes based on input text prompts, making it a specialist for 3D interiors. Apple demonstrated that GAUDI could randomly generate new camera movements via 3D scenes. It could start from an image or prompt with an input like, “go down the stair.”
Excited for this to be out! Introducing GAUDI: a generative model for 3D indoor scenes. We tackle the problem of learning a generative model of 3D scenes parametrized as radiance fields. This has been a great collaboration across multiple teams at @Apple. https://t.co/aJOqtzA2CIhttps://t.co/tSkJdXK31Cpic.twitter.com/ReeXAPGg95
— Miguel Angel Bautista (@itsbautistam) July 29, 2022
Neural rendering combines artificial intelligence with computer graphics. The technology of Neural Radiance Fields (NeRFs) has been utilized as a neural storage medium for 3D scenes and models. These models can be rendered via different camera angles.
AI systems based on this technology show the potential of controllable generative AI, but only for two-dimensional graphics. The limitation stems from the limited possibility of camera positions. Camera positions are restricted by obstacles like walls and objects when rendered in 3D.
Apple’s GAUDI model solves this problem with its three-tier specialized network. The network consists of a camera pose decoder (for predicting possible camera positions), a scene decoder (for predicting a tri-plane representation), and a radiance field decoder (for drawing the following image via volumetric rendering equation).
GAUDI’s video generation quality is yet to improve as it is still filled with artifacts. Apple is constantly working on its AI system and laying more foundations for generative AI to render #D objects and scenes.
ARTPARK, a Bangalore-based AI and robotics foundation, has announced a $100M venture fund to invest in 10-15 growth-stage AI and robotics startups that develop products for education, mobility, and healthcare sectors. The venture fund will support the startups over the next three years.
The Securities and Exchange Board of India (SEBI) has approved the funding, and it will make its first closure this year by December. ARTPARK will invest in new as well as existing portfolio startups.
ARTPARK cofounder Umakant Soni said that the foundation understands the phase where entrepreneurs need financial backing since they are entrepreneurs and have built technologies by themselves. ARTPARK is offering startups the required risk capital, he added.
Apart from launching a venture fund, ARTPARK has plans to support nearly 52 AI and robotics startups in India over the next five years. ARTPARK aims that the funded startups gain unicorn status and turn into decacorns to hold the potential of becoming the change that can reshape the entire business ecosystem, Soni said.
The AI & Robotics Technology Park (ARTPARK) is a not-for-profit foundation founded in 2020. It is a brainchild of the Indian Institute of Science, Bengaluru, and venture studio AI Foundry to promote technological innovations in the sector.
ARTPARK aims to support AI and robotics startups fostering innovations in various fields such as healthcare, education, agriculture, infrastructure, agriculture, mobility, retail, and cybersecurity. The foundation has invested ₹75 lacs each in 19 startups in the AI and robotics space so far.
Despite recent scandals over data privacy and misinformation, Meta’s sales have kept on growing without fail throughout the years. However, the social media conglomerate on Wednesday reported a 1% decline in second quarterly revenue from the previous year. This is the first time Meta’s revenue has fallen since it went public a decade ago.
Meta’s revenue for the second quarter decreased from $29.07 billion a year earlier to $28.82 billion. Profit was down 36% from a year earlier to $6.69 billion. Meta’s declining revenue was astounding, given that its quarterly revenue growth was up 28% during 2019.
Given that, questions are being asked. What are the possible reasons behind the sudden revenue decline? And most importantly, will Meta recover from the slump? Let’s consider what we know.
According to Meta, broader economic uncertainty and weaker demand for digital advertising is the reason for the revenue decline. Zuckerberg said that we have entered an economic downturn that will have a long-term impact on Meta’s digital advertising business.
Companies like Google, Twitter, and Snap, which rely on online advertising, have also experienced reduced demand for advertising due to the slowing global economy. Some companies blame the destabilizing effect of the Russia-Ukraine war on the European advertising market and the strength of the US dollar. These factors are the reason for decline in global sales and revenue for several advertising-based companies. E-commerce ads started witnessing a decline as the peak pandemic passed and more people ventured outside, resulting in challenging periods like these.
One of the main reasons for Meta’s revenue decline is related to some changes made by Apple. Last year, Apple employed some privacy-related changes that disabled Meta from measuring and delivering its advertising on Apple-made mobile devices. This is a disadvantage as Meta makes the vast majority of its advertising revenue from smartphone devices.
Meanwhile, Zuckerberg has been making substantial financial investments toward his vision of the metaverse. According to him, the effort may take years to come to fruition, and the endeavor will be costly. However, some investors are now skeptical about whether the investments will pay off in the long run.
Will Meta make it through?
The rundown is unlikely to end soon, as Meta said it expects a continuation of the weak advertising demand environment for the current quarter. However, Meta is now shifting its focus to TikTok, the Chinese short video app that has become immensely popular in just a few years. Zuckerberg has begun to shift the company’s focus to video products, namely Reels, that are driving growth and engagement and is making sweeping changes to Instagram and Facebook.
Zuckerberg aims to eventually make more money from Reels (short videos) as an advertising asset, which currently is not as engaging on Instagram as its counterpart TikTok. Considering the increasing figures for the demand for short video content and the already engaged audience for Reels, it is most likely that Zuckerberg’s plan would attract more revenue. Besides, the recent investments in artificial intelligence recommendation algorithms are also encouraging more people to use the Reels for advertising for extended periods.
Besides the lackluster financial results, Meta’s earnings report also had some bright spots. According to the report, the daily active users on its family of apps, including Facebook, Instagram, and WhatsApp, rose by 4% from a year ago to 2.88 billion. These results exceeded the predictions of trend analysts that Meta will lose users in light of the revenue decline. Moreover, the Facebook app witnessed incredible user growth in the US.
Conclusion
After considering all the above information, the recent revenue decline seems like a temporary hiccup. According to industry experts, Meta has enough resources at its disposal to overcome this period of economic rundown and transition into its usual profitable self. To make it through a difficult period, the company has slowed down hiring in the year’s second half and has cut costs. Meta also said that the chief financial officer, David Wehner, will now serve as chief strategy officer to help with challenges, partnerships, and internal organization.
Cryptocurrencies, including the reputable Bitcoin, have been experiencing a slump in the stock market and other investments. Crypto experts are calling this bear market for cryptocurrency a crypto winter. But like a bear market for stocks that keeps recovering, will this crypto winter end?
How bad is the cryptocurrency crash?
According to Coinbase, Between July 2021 and July 2022, Bitcoin has fallen by about 24%. The current value of Bitcoin is $21,070.10 per coin as of July 26, roughly 33% less than its price in December 2021 of $64,912.20. Similarly, Ethereum is down more than 13% from this time last year, and it has lost more than 32% of its value since its peak. Altcoins usually tend to follow the pricing trends of Bitcoin, and according to Forbes Advisor, the global crypto market has dropped about 60% since November of last year.
However, following the same tracks as the stock market, which saw some modest gains in July, crypto is also creeping up at a decent pace. Coinbase said Bitcoin showed a 21% increase in the second week of July. But that does not necessarily mean that the crypto crash is over. But, let us first consider whether the crypto market will be able to survive the recent crash.
When the cryptocurrency market was declining, Terra was supposed to stay put as it was a stablecoin, i.e., its value is tied to the value of the US dollar. Ironically, Terra dropped from more than $50 billion to virtually worthless in one week. Following this, other stablecoins also began to lose value. Soon after, Bitcoin started to drop as well. While the failure of Terra was one of the catalyzing factors, other economic factors, including rising interest rates and inflation, also contributed to the crypto crash.
Crypto experts say that many coins will cease to exist at the end of the latest crypto winter. For example, many penny cryptos, which are highly volatile investments, may crash with no hope of revival whatsoever. However, the blue-chip cryptocurrencies, which include Bitcoin, are more likely to survive the current cryptocurrency market crash. Regardless of price, cryptocurrencies in the top 500 by market cap are more likely to give positive returns over time.
Cryptocurrency’s viability is also tied to the technology that powers it, i.e., blockchain. With the growth of the metaverse, especially NFTs and NFT gaming, the technology behind cryptocurrency seems more prevalent and promising than ever. The underlying features of blockchain technology represent huge improvements for commerce as businesses worldwide using the tech will be able to operate much cheaper and faster with greater transparency and security.
Is the Crypto Market Recovering?
The industry has seen a recent rise of crypto in mid-July and other positive economic indicators that indicate a slow turning of the bear market. These changes may perhaps be indicating a cryptocurrency market recovery over the horizon. And it may even happen faster than many experts believe it would occur.
There was an unexpected boost in retail spending in June 2022 in the US. Also, the US Federal Reserve projected an interest rate hike of just 0.75% rather than a whole point. These factors have bode well for the US economy. And since the crypto crash is related to these kinds of macroeconomic factors related to countries like the US, crypto could, in fact, be in the early stages of recovering already.
Conclusion
There have been instances where the cryptocurrency market has bounced back before. And if the stock market is referenced as an example, it is likely that the crypto market will survive and rebound again soon. However, will crypto ever go back up to its 2021 record highs? Some experts believe not only will crypto bounce back up but that Bitcoin could surpass $100,000.
Investors should stick to coins with a known reputation and a high market cap, such as Bitcoin, to increase the odds of investment success.
Several organizations, including those in the legal industry, use blockchain technology for various business functions. Blockchain enables users to record transactions over a distributed computer network. Since the server is secure and the transactions are permanent, the verification is more straightforward. However, as blockchain processes are steadily becoming a part of more and more financial lives, people are asking whether the blockchain can be hacked.
How safe is blockchain?
According to experts, the blockchain itself cannot be hacked; however, blockchain-adjacent processes can, and that too in several ways. That means that blockchain transactions can be manipulated, and blockchain assets can be stolen. Nonetheless, this is not the reality of blockchain itself. Instead, it is more about the environment in which blockchain assets are owned and traded.
Most of the so-called ‘blockchain hacks’ that have happened in the past few years have been on centralized exchanges. In some situations, one has to use an exchange to trade blockchain assets or cryptocurrency. However, hackers can access digital assets through an exchange platform or network. In simple terms, if we consider the example of Bitcoin, there is no central system to hack because it is naturally decentralized. Exchanges, however, put the assets into a ‘place,’ and those places can be exposed to hackers.
There are also instances where some hackers sniff out a vulnerability in an exchange and make off with someone else’s assets. ‘Rug pulls’ are instances where someone gets others to invest in an asset and then take off with their money. However, it must be noted that none of this happens in the blockchain itself.
The 51% Attack
In a given blockchain, the community of owners supports the integrity of network transactions. For example, Bitcoin ownership gets verified using the blockchain ledger, through the consensus of the entire community of Bitcoin owners. If one party manages to gain control of more than 50% of that ownership, then everything related to the blockchain transactions can be manipulated by them. This is called the 51% attack as the accomplishing party is the majority owner, and they have a say in what happens.
It is challenging to execute a 51% attack in reality. It is prohibitive in a network of any size. In a practical sense, it is generally not possible for anyone to own 51% of Bitcoin or Ethereum or any of the other significant blockchain assets.
Smart Contracts
During the past couple of years, new advancements have occurred in the blockchain security world, including the introduction of smart contracts. Smart contracts allow putting data and code executions on the blockchain. They can be considered non-financial blockchain transaction vehicles. Smart contracts started getting popular as users began investing more in cryptocurrency.
According to IBM, one of the benefits of smart contracts is that through them, blockchain transaction records are encrypted, thus making them very hard to hack. Moreover, hackers would have to alter the entire chain to alter a single record as each record is connected to the previous records and the subsequent ones on a distributed ledger.
To conclude, smart contracts will have to be hacked in ways that cryptocurrencies cannot. And if a hacker exploits some aspect of the smart contract that is blockchain-adjacent, it can look like the blockchain is hacked, which is not true.
Conclusion
It is a fact that the blockchain itself as a model is very resistant to almost all kinds of hacking. However, many systems and processes connected to a blockchain and assets have vulnerabilities and can be compromised. This is crucial to be kept in mind as the market continues to see more kinds of crypto coins and smart contracts develop in the constantly expanding network of fintech.
A large language model called BLOOM (BigScience Large Open-science Open-access Multilingual Language Model) with 176 billion parameters, was released by the BigScience project. On June 17, 2022, a preliminary version of the BLOOM language model was made available. The Bloom language model, which was created with the assistance of around 1,000 academics and researchers from across the world to challenge big tech’s dominance over large language models, will be open source and the first model of its magnitude to be multilingual.
The BigScience research project was initiated in 2021. It comprises over 250 institutions and researchers from more than 60 countries. While Hugging Face is the principal investigator on the project, it also includes researchers from GENCI, the IDRIS team at the CNRS, the Megatron team at NVIDIA, and the Deepspeed team at Microsoft.
How is BLOOM competing with other large language models?
The researchers explain in their paper that large language models are algorithms that learn statistical correlations between billions of words and phrases to accomplish tasks including creating summaries, translating, answering queries, and categorizing material. The models are trained by tweaking values, referred to as parameters, by redacting words and comparing their predictions with reality. BLOOM contains 176 billion parameters, matching one of the most well-known models of its kind, GPT-3, developed by the nonprofit OpenAI and licensed by Microsoft.
Despite the impressive accomplishments these large language models have produced (such as writing articles), they cannot comprehend the fundamentals of human communication and language, leading them to produce nonsensical output. The fact that they might encourage abuse or self-harm and reflect pre-existing racial or gendered stereotypes woven into the human-written content they learn from is even more concerning. Furthermore, training these models typically costs millions of dollars and has a massive carbon footprint.
Thorough knowledge of how these language models are created, how they work, and how the larger community can improve them is essential, considering the possible influence of such language models. Popular models, like GPT-3, are not available as open source. This indicates that only a small number of individuals are aware of how these models work internally. Most large technology companies creating cutting-edge large language models prohibit others from using them and keep their models’ inner workings secret. It is challenging to hold them responsible because of this.
The main inspiration behind BLOOM was to challenge and change these norms of opacity and exclusivity!
BLOOM was created by hundreds of academics, including philosophers, lawyers, and ethical experts, in addition to staff members from Facebook and Google, unlike previous large language models. A US$7 million investment in public computing time is being used to train BLOOM. BigScience was given free access to France’s national Jean Zay supercomputer (IDRIS) facility outside of Paris in order to train BLOOM.
To fully utilize the computational power available, the researchers polished up the data collection using a multilingual web crawl, vetted for quality and with minor redaction for privacy. The team also made an effort to lessen the typical over-representation of porn sites, which might cause sexist connotations in the model, without eliminating keywords that would exclude information related to open discussions of sexuality in frequently under-represented communities.
Despite the aforesaid precautions, researchers acknowledge that BLOOM will not entirely be bias-free, but they expect to advance current models by supplying it with diverse and excellent sources. Importantly, since the model’s code and data collection are public, researchers could try to identify the causes of undesirable behaviors, which could enhance subsequent versions. In addition, BLOOM attempts to disrupt the sway of major businesses over large language models. It achieved that since the project was created in an open environment and makes use of an open license based on the Responsible AI license. BigScience created this license to discourage the use of BLOOM in high-risk industries like law enforcement or health care, as well as to harm, defraud, exploit, or mimic individuals. According to Danish Contractor, an AI researcher who volunteered for the project and co-created the license, the license is an experiment in self-regulating large language models before laws catch up.
Addressing Availability
Bloom is capable of understanding texts in 46 native languages and dialects, as well as 13 computer languages. The native languages include French, Vietnamese, Mandarin, Indonesian, Catalan, 13 Indic languages (such as Hindi), and 20 African languages. Only little over 30% of its training data was in English – thus making it an exception from large language models, where English dominates.
Bloom can be tasked with creating summaries or translations of text, output code from instructions, and follow prompts to complete original tasks like writing recipes, extracting data from news articles, or constructing sentences using a newly-defined invented word, despite the fact that it was never trained on any of those particular tasks.
For researchers who want to experiment with it or train it on fresh data for particular applications, the fully trained BLOOM model has been made accessible for download. However, downloading and using it call for a sizable amount of hardware. BigScience has also provided scaled-down, less resource-intensive versions of the model as well as developed a distributed system that will enable laboratories to share it across several servers. Hugging Face has even released a web application that will allow anybody to query BLOOM without installing it.
Wrapping Up
Large language models are one of the most exciting and hottest topic of research in the AI industry. As this trend dominates the sector, companies are racing to build a larger (in terms of parameters) and more capable model. Cerebras Systems said last month that it has achieved a record for the biggest AI models ever trained on a single device, in this instance a massive silicon wafer with hundreds of thousands of cores.
While some businesses have chosen to compromise on the unfairness and privacy loss posed by large-scale language models, others have chosen to open source some of their language models, such as Yandex’s YaLM 100B.
At the same time, experts are questioning the use of enormous datasets and computing power by DeepMind’s Gopher and Chinchilla models, OpenAI’s GPT-3, Google’s LaMDA and PaLM, and DeepMind.
While BLOOM claims to address all these concerns, it also needs to improve on its performance before going mainstream.
Under the collaboration, Vehant Technologies will sponsor research projects and consultancies at IIIT Delhi. The company will also support MTech and Ph.D. students, starting with sponsoring up to five MTech students for the academic session 2022-23. The projects are all set to commence in August.
Through the sponsorship, students will receive full coverage of the cost of their education, including reimbursement of living costs. They will also get first-hand industry experience on live projects while simultaneously continuing their regular academics.
The IIIT-Delhi MTech students receiving the sponsorships will be called Vehant fellows. Each fellowship will have a commitment of Rs.10 lakh from Vehant Technologies throughout students’ academic courses, i.e., two years. Students can flexibly utilize the fund for research and educational purposes.
The Institute said, while IIIT-Delhi researchers and faculty are well equipped with experience in the field of computer science and engineering, the collaboration with Vehant Technologies will further propagate the system knowledge of artificial intelligence and machine learning among students.
According to an IBM Cost of Data Breach Report 2022, which calls cyberattacks the biggest challenge to the industry, data breaches have cost Indian businesses an average of ₹17.6 crore in 2022, the highest amount ever recorded.
The cost increased 6.6% from last year, when the average breach cost was Rs 16.5 crore. The IBM report said the cost was up 25% from ₹14 crore in 2020.
Industrial companies, including engineering, chemical processing, and manufacturing, paid the highest for data breaches. The average cost of each breach was about ₹9,024 this year.
For the services industry, including legal, accounting, and consultancy, the average cost of a breach was ₹7,085, and for technology industries comprising software and hardware companies, the cost was ₹6,900.
Costs incurred after a security breach was the largest at ₹71 million for the sixth year, among four categories of lost business, notification, post-breach response, and detection and escalation. According to the report, the global average data breach cost reached an all-time high of US $4.35 million for surveyed organizations.
According to the report, keeping security capabilities flexible enough to match attacker agility will be the biggest challenge as the industry moves forward. To stay on top of growing cybersecurity challenges, investment in mature security practices, zero-trust deployments, and AI-based platforms can help, the report added.
Facebook has reported its first-ever yearly decline in revenue for the second quarter, witnessing a 1% drop which left it at $28.8 billion. It is predicted that growth in the third quarter could fall even more. The overall profit for its parent company, Meta, fell by 36% to $6.7 billion.
The Reality Labs division responsible for building Mark Zuckerberg’sZuckerberg’s metaverse dreams lost $2.8 billion in the quarter. Suddenly, Meta’s business has become challenged on all fronts.
The reason behind the decline is Apple’s ‘‘Ask app not to track’’ prompt. The prompt on iPhones has made Meta’s ads much less effective, costing the company $10 billion in ad revenue last year alone. Now, a rapidly slowing economy has caused advertisers to pull back on their spending.
Meta, to compete with TikTok, is rearchitecting Instagram and Facebook to place emphasis on short videos and posts that its system recommends to people.
According to Zuckerberg, the company had seen stronger than anticipated engagement trends on Facebook due to an increase in the consumption of videos. In the long run, the company expects Reels to be a revenue driver. However, even though the company is prioritizing reels, it is not making much money from them.
Despite the declining revenue, Meta has managed to grow Facebook’s daily users by 3% to 1.97 billion. It comes after an alarming user decline observed a couple of quarters ago. Meta reported that a total of 2.88 billion people now use its suite of social apps, including Facebook, Messenger, Instagram, and WhatsApp, which is an increase of 4 percent from a year ago.
