Scientific Program | Artificial Intelligence | Machine learning

Call to accelerate AI and robotics adoption to combat rising labour costs

2025-05-02 - 2025-05

Companies must immediately increase their investments in AI and robots to guarantee a robust and prosperous food system, while addressing a burgeoning population and rising labour expenses.

According to James Walton, chief economist at IGD, an expanding population, tariffs, climate change, and rising labour costs are all severely straining the resilience of the food system.

The latest IGD Economic Viewpoint Report, “How to respond to rising labour costs in 2025”, released on April 29th, highlights a perfect storm brewing in the food system. Rising labour costs, geopolitical risks, and inflation are creating a challenging environment for businesses and consumers alike. This combination of factors is leading to a highly volatile and uncertain environment, impacting both consumer and business confidence, and affecting operations across the food system.

In its key findings, IGD forecasts retail food prices to rise 3.4% on average compared to 5.2% (source IGD research, Retail November 2024; Away From Home April 2025) for away-from-home (AFH) food prices in 2025, fuelled by rising labour costs.

Financial confidence is down sharply, with only 23% expecting to be better off in the next year, the lowest since November 2023.(source: IGD ShopperVista research, March 2025; Base 1,000+ adults aged 18+)
Over eight in ten consumers expect rising prices for food, grocery, and dining out (source: IGD research, 21-23 March 2025, 1085 adults aged 18+).
Retailers are increasingly turning to technology to boost productivity, tackle staffing gaps and cope with rising labour expenses.
Michael Freedman, head of economic and consumer insight, summarises the report’s findings: “The IGD Economic Viewpoint highlights the multifaceted challenges facing the food industry, from rising labour costs and geopolitical risks to inflation and dipping business and consumer confidence. The report underscores the need for strategic resilience, innovation, and collaboration across the food system.”

The food industry has seen nearly 200,000 fewer people working in food manufacturing since 1980, with the number dropping from 613,750 in 1980 to 415,750 in 2024. Despite a population increase of 13 million people in the UK during this period, food manufacturing has become more labour-efficient, with each worker now serving 167 people compared to just 91 in 1980 (source ONS Data 2025).

April 2025 marks a critical point as measures introduced in the Autumn Budget 2024, including the higher National Living Wage and changes to employer National Insurance Contributions, have sharply increased labour costs, particularly for part-time and lower-wage roles. With profits already thin, there is no room to absorb these cost increases, and consumers are likely to end up paying more.

With the UK population set to grow by 7 million in the next 20 years (76.1 million by 2045 vs. 69.2 million in 2024), the focus must shift. The food industry faces two significant challenges: dealing with rising labour costs and the need to feed more people.

Expanding the capacity to produce, process, and distribute food domestically is essential and the IGD said industry needs to embrace the next wave of technological advancements, particularly in AI and robotics, to meet these challenges.

Despite the challenges, there are opportunities for the entire food industry to adapt. Retailers are increasingly turning to technology to boost productivity, cut costs, and cope with rising labour expenses amidst staffing challenges. In retail these advancements are reshaping efficiency and setting new benchmarks. Beyond daily tasks, technology is transforming inventory management, reducing theft, and unlocking revenue through data monetisation. In a rapidly evolving market, technology provides retailers with the edge they need to adapt and succeed.

Toby Pickard, retail futures senior partner, emphasises the urgent need for food businesses to build a better system through automation and innovation. “The food industry is at a critical juncture. By leveraging technology and innovative solutions, we can transform the sector, ensuring sustainability and growth. It’s not just about reducing headcount; it’s about creating a resilient and efficient food system that can weather these challenges and emerge stronger.”

AI VALLEY: Microsoft, Google, and the trillion-dollar race to cash in on Artificial Intelligence

2025-04-24 - 2025-04

‘It played like God’, said Garry Kasparov, the world chess champion, as he accepted defeat from IBM’s supercomputer Deep Blue in a man vs machine match in 1997. Kasparov contemplated that if he was playing a human, he would have thought of a few anticipatory moves and used his speed to create tension, forcing his opponent to make mistakes. But the emotionless machine had rapidly calculated many moves ahead and played with such tenacity that it unnerved him enough to surrender the game. Kasparov’s defeat marked a major milestone in machine intelligence.

I remember the sci-fi fantasy movie Her (2013), in which a lonely man falls in love with his operating system. It becomes his constant companion, sensing his needs and moods and conversing like a human through the day. That science fiction is now real. Launched in early 2023, the AI application, Pi, enables users to create their own personal assistants with the mission to make them happier, healthier and more productive.

This obsession of programmers to teach machines to replicate human behavior has led to an unprecedented revolution in technology. Pulitzer Prize winning author, Gary Rivlin’s new book AI Valley chronicles the evolution of computing from the early days of the invention of silicon chips, the earliest programming software to the remarkable race being played out right now to win the artificial intelligence battle with trillions of dollars at stake.

In the form of a page turning thriller, Rivlin’s book details how AI has rapidly permeated everyone’s lives. The iPhone arranges the photographs you click into albums on its own, word processors correct your grammar, write letters for you, programmes do homework, answer your phone and pay your bills. Generative AI apps like ChatGPT can now make your still photos dance and even convert screenplays into high-quality films.

If you had to only read one book about the rapid advances of technology, this is it. Rivlin has covered the history of nearly every player in the race to the top of the game, including the VCs who funded and the companies they funded. Companies tumble out of the book at the pace of nearly a company a page, covering Microsoft, Intel, Facebook, Airbnb, Uber, Google and hundreds of startups, including newer AI ones like DeepMind, OpenAI and Nvidia.

Microsoft’s big comeback
The biggest hero of this book is easily Satya Nadella. Microsoft, the earliest mover in the computing space was left out when FAANG (Facebook, Apple, Amazon, Netflix and Google) became the new darlings at the stock market. After two decades of no growth, missing the search, mobile computing and social tech revolutions, Nadella rapidly shifted the focus of the organisation to cloud computing and AI when he took over as CEO. With investments in the two largest AI companies OpenAI and Inflection AI, Microsoft swiftly multiplied revenues and within a few years, at $3trn in market capitalisation, is back as the hot stock ahead of Google ($1.8trn), Amazon ($1.6trn), Facebook ($1trn) and Netflix($0.5trn).

Rivlin uses multiple data points to show how Generative Ai is not like a conventional start-up investment due to the large-scale funding needed in mining huge amounts of data, high speed processing computing power and expensive tech talent that just doesn’t get excited with only stock. Even with such high entry barriers, GenAI has managed to raise seed funding of billions of dollars.

If this book was a film, it would be the story of Reid Hoffman, a former product manager at Apple, called the Midas VC in Silicon Valley, for his extraordinary ability to seedfund some of the biggest tech companies of our times and of Suleyman, his protégé, the Syrian whizkid who formed two of the biggest AI companies, DeepMind and Inflection AI. While Hoffman is the founder of LinkedIn, he’s more famous for his investments in companies like PayPal, Facebook, Airbnb, Uber, Flikr, Groupon, Convoy and OpenAI and has used his extensive networking skills to build these companies to great scale.

Rivlin says that Hoffman used to brief the Pope at the Vatican annually on how AI will change the world in education, medicine, agriculture and living. DeepMind, for instance, made the most meaningful AI breakthrough in 2022 when it discovered AlphaFold which gave humans an understanding of protein structures that would help new drug discoveries and therapies in pursuit of immortality.

The book balances the narrative with the pitfalls and challenges of too much AI, of computers suffering hallucinations, or spewing irrelevant answers occasionally. AI machines need to be continually trained by humans to respect racist, sexist, and religious sentiments as Rivlin warns of the dangers of unregulated AI, with its abilities to create a pandemic or even empty out a bank.

The Scramble for the top
AI is not a love affair, it’s a rage, says Rivlin, as he talks of how every VC is hunting for an AI idea to fund and how the companies they fund scramble for supremacy. US politicians have always tried to rein in big-tech and just like the speed of adoption of technology, the call from Congress to testify comes faster too. In 1999, after nearly two decades selling software, Bill Gates was summoned to explain his predatory practice of bunding his software for competitive dominance and threatened to break his company. In under a decade, Facebook was called in on data privacy, but Sam Altman of OpenAI was called in barely within his first year to explain how machines will be stopped from doing evil.

When it comes to technology, time and again, it’s been proven how the current leaders of tech are humbled by the new. For instance, the early chatbots Siri and Alexa did not try to improvise like ChatGPT whose strikingly human qualities like the ability to tell a joke, admit mistakes, and ask follow-up questions fetched it 900 million downloads in nine weeks. ChatGPT also leapfrogged Google search and provided direct answers instead of 10 blue ticks for linked articles.

Artificial Intelligence (AI) and Machine Learning Market to Witness Groundbreaking Innovations by 2025 - 2032 | Qualcomm, Intel, Google, Ward Systems

2025-04-02 - 2025-04

The Artificial Intelligence (AI) and Machine Learning Market was valued at $120.9 Billion in 2022, and is projected to reach $190.5 Billion by 2032, growing at a CAGR of 32% from 2023 to 2032. According to the latest research from Coherent Market Insights, the Artificial Intelligence (AI) and Machine Learning Market is projected to experience significant growth between 2025 and 2032. This market intelligence report offers in-depth analysis based on thorough research, highlighting current trends, financial performance, and historical data evaluation. The company profiles within the report are derived from the current performance of the Artificial Intelligence (AI) and Machine Learning Market, considering key factors such as drivers, trends, and challenges, as well as global market share, size, and revenue forecasts for comprehensive insights. To provide a clear understanding, the report examines leading companies, types, applications, and the factors contributing to a positive future outlook. The Artificial Intelligence (AI) and Machine Learning market report includes detailed charts, tables, and data analysis, with clear objectives aimed at potential stakeholders. It offers a comprehensive study of the Artificial Intelligence (AI) and Machine Learning market, providing valuable insights to support strong growth in the future. This report is designed for a wide range of interested parties, including stakeholders, market participants, investors, researchers, and other individuals associated with the business.
The Global Artificial Intelligence (AI) and Machine Learning Market Analysis Report offers a comprehensive overview of the market size across various segments and countries, including historical data and future forecasts. The report outlines the competitive landscape of the global market, discussing market dynamics, drivers, and segmentation by application, type, region, and manufacturer. It provides both qualitative and quantitative insights into the industry for the regions and countries covered. Additionally, the report highlights key opportunities in the Artificial Intelligence (AI) and Machine Learning market, identifying the factors driving growth and expected to continue fuelling expansion. It also considers past growth trends, current drivers, and future market developments.

The report provides market size (2019 - 2031) Value (US$ Mn), Y-o-Y % growth rate, and forecast (2024 - 2031), CAGR % for all the segments and sub segments wherein:

• 2020-2023: Historic year,
• 2024: Base Year,
• 2025: Estimated Year,
• 2025 to 2032: Forecast Period.

📌 Key Highlights of our Artificial Intelligence (AI) and Machine Learning Market Research Report:

» Comprehensive analysis of the Artificial Intelligence (AI) and Machine Learning Market
» Identification of market size and growth trend
» Competitive landscape assessment, including key players and their strategies
» Consumer behaviour insights related to Artificial Intelligence (AI) and Machine Learning usage
» Emerging trends and opportunities in the Artificial Intelligence (AI) and Machine Learning Market
» Regional analysis, highlighting variations in Artificial Intelligence (AI) and Machine Learning usage and competition
» Industry best practices for effective Artificial Intelligence (AI) and Machine Learning optimization
» Future outlook and market projections for informed decision-making

How to create free Ghibli-style AI images using Grok with help from ChatGPT: Check our step-by-step guide

2025-03-31 - 2025-03

ChatGPT's new native image generation capabilities continue to wreak havoc on OpenAI's servers as users clamour for the chatbot to generate and convert their real life images in Studio Ghibli style. While ChatGPT's native image generation capabilities allow it to create more detailed and contextual images than other chatbots, it is also incredibly limited due to high demand, with free users currently only allowed to create three images, while even paid users have a usage limit. xAI's Grok chatbot, on the other hand, may not be the most accurate at creating images, but generally has a longer image upload and generation limit (no specific limit given by the company). But what if you could use ChatGPT to create more nuanced images using Grok? Check out the step-by-step guide below. How ChatGPT can help create free and better images using Grok? While modern large language models (LLMs) can understand natural conversational tones to generate images, these images can often be disappointing as users miss certain details or the chatbot hallucinates (makes up) key elements of the images. This is where it is important to generate detailed prompts, taking into account context, subject, background, theme, colour palette, atmosphere, and art style. Avoiding ambiguity is paramount here, as this would drive the chatbot to generate unexpected elements in the image. This is where the use of a chatbot such as ChatGPT or even Gemini can prove useful to provide detailed instructions that take into account the user's preferences while avoiding ambiguity in the images. We tried using Grok to create a Studio Ghibli-style image of three popular Indian captains, but the end result was disappointing. However, things were very different when we gave ChatGPT the initial prompt and asked the chatbot to create a text prompt to create the image using Grok. As you can see from the two images below, the first image gets most of the details wrong, with faces that bear no resemblance to the original players and the wrong Royal Challengers Bengaluru jersey. Even the background in this image looks unconvincing. Contrast this with the second image, generated using text prompts from ChatGPT, where the players' faces, the Ghibli-style effect and even the pattern of their jerseys are mostly correct. The image generated by Grok still got the names of the franchises and their sponsors wrong, but the team logos were accurate.

How Adobe Acrobat AI Assistant improves accuracy and efficiency in contract review

2025-03-28 - 2025-03

It doesn’t matter if you are a large company, a self-employed worker, or an ordinary user; in this era, marked by the technological revolution, we all seek new tools that can help us improve and optimize our productivity. Likewise, there is a very specific area that can be (and, in fact, has been) benefited by the latest technological innovations: contract review.
Reading a contract carefully is a task as tedious as it is essential. Fortunately, AI Assistant, one of Adobe Acrobat’s latest features, can be very helpful.

What is AI Assistant?
The Adobe Acrobat AI Assistant, also known as AI Assistant, is a tool powered by artificial intelligence designed to facilitate, among other things, the review and analysis of your documents.
AI Assistant is integrated within Adobe Acrobat and can identify, highlight, and suggest changes in your documents, such as a contract. This way, you can also be sure that you won’t overlook the most important details.
That said, before we get to work, there are several sections you should keep in mind:
• The languages: currently, AI Assistant can analyze files in English, French, German, Spanish, Italian, Brazilian Portuguese, and Japanese.
• The size: your document cannot exceed 100 MB and must have fewer than 600 pages.
• Being connected to the Internet: AI features must be enabled within Adobe Acrobat and you must remain online at all times.
Don’t let anything slip away: detect errors, omissions, and much more
One of the biggest challenges we face when reviewing a contract is understanding its content. AI Assistant can scan the document and summarize the most important parts for you, so you always know what you will sign.
• To start, open your PDF file with Adobe Acrobat or drag it into the program once it’s open.
• Click on the Generative Summary icon, which you will find under the AI Assistant button. If you don’t see it, select All Tools and search for Generative Summary.

Salesforce and Deloitte Strengthen Their Partnership to Apply AI Agents to “Every” Enterprise Process

2025-03-27 - 2025-03

Deloitte and Salesforce have strengthened their decades-long partnership, striving to deliver scalable agentic AI solutions across multiple industries. As part of the announcement, Deloitte presented its new marketing agent, powered by Salesforce’s Agentforce. The “master” agent supports marketing teams with many use cases for optimizing customer journeys, from generating briefs to pinpointing customer journey improvement opportunities. Yet, the partnership is not only about delivering this offering across the marketing space; it’s much broader. Indeed, Deloitte and Salesforce are teaming up to create a burgeoning suite of industry-specific agents and accelerators that cross various enterprise functions and verticals. Regarding the latter, the partnership targets financial services, government and public services, life sciences, and healthcare. By applying Deloitte’s professional services and Salesforce’s platform – which also features Data Cloud and its Customer 360 Apps – businesses can “achieve significant efficiency gains and competitive advantages.” Speaking on the enhanced collaboration, Brian Millham, President & COO of Salesforce, explained: “AI agents are a massive opportunity for organizations to unlock an unlimited digital workforce that collaborates with humans to increase productivity, reduce costs, and streamline operations. Agentic AI Hype Is Accelerating, So What’s the Hold Up for End Users?
While GenAI was the talk of the town last year, agentic AI is for sure one of the hottest topics across the CX landscape. Yet, considerable levels of hype seem to be coming from the vendor’s side over end users. As the landscape of Agentic AI continues to evolve, enterprises are at various readiness levels. Indeed, some have concerns with how they track and manage their data. After all, if they have misconfigured permission settings, the AI agent may accidentally surface confidential financial information. Meanwhile, others worry about onboarding the necessary technical skills and how AI agents may change employee roles. Nonetheless, there are companies matching this hype set by the vendors.
Speaking of the unique benefits for law firms, Lindsay Marty, Founder and CEO of Above the Bar Marketing, told CX Today: “A lot of our pain points are around missed opportunities—leads that come in after hours or responses that fall through the cracks. AI can plug that gap.

How DataRobot is Pushing AI Use Cases to Production

2025-03-26 - 2025-03

Headquartered in Boston, enterprise AI company DataRobot recently unveiled its enterprise AI suite to simplify the creation and deployment of AI applications. The AI suite helps develop and deliver generative AI applications and agents. A few weeks prior to the launch, AIM got in touch with CEO Debanjan Saha, who discussed the company’s strategy of addressing three critical gaps hindering widespread AI adoption in enterprises: the value gap, the confidence gap, and the expertise gap.
Addressing the Value Gap
Saha emphasised the need for AI investments to deliver tangible business value. “People are building data centres. At various conferences, you will hear that people are building gigawatts of capacity,” Saha noted. However, he stressed that these investments must translate into solving real business problems beyond creating simple chatbots. To bridge this gap, DataRobot is focusing on helping customers identify and implement AI use cases that drive significant business impact. “Often, in order to take an AI use case to production, from ideation to final stage, there are 19 different teams who have to participate in a collaborative way to make it through that process,” Saha revealed, highlighting the complexity of implementing AI solutions at scale. The confidence gap, as described by Saha, stems from various risks associated with AI deployment, including operational, accuracy, reputational, and regulatory risks. Due to these concerns, many organisations are hesitant to move beyond prototypes and demos. DataRobot has implemented robust governance and monitoring features in its enterprise AI suite to address this gap. For regulated industries such as financial services, the company provides comprehensive AI governance, such as AI observability with real-time intervention and moderation and one-click compliance documentation to empower businesses to deploy AI with confidence. “Before AI can go into production, it has to go through a set of checks, and a compliance report needs to be produced for both internal auditors and external regulatory bodies,” Saha explained. This approach has made DataRobot particularly attractive to financial institutions, with 60% of the top banks in the United States using the enterprise AI suite. Having recognised the scarcity of AI talent, especially in enterprises, DataRobot is working to make AI more accessible to a broader range of professionals. The company’s strategy involves both technological solutions and hands-on support. On the tech front, the enterprise AI suite aims to lower the bar for participation through automation and user-friendly interfaces. Additionally, the company offers executive sessions, ideation workshops, and supervised hackathons. “We have a team of highly-skilled data scientists who have years of experience working with customers even before generative AI to help solve business use cases with AI,” Saha said. These specialists, known as applied AI experts, work alongside customer teams to upskill and assist them in implementing AI projects.

Beyond the AI hype: Keep tabs on the AI tools used in the cloud

2025-03-24 - 2025-03

AI, for all its intelligence, is not risk-free and requires your attention. Cloud-based AI is prone to avoidable toxic combinations that leave sensitive AI data and models vulnerable to manipulation, data tampering and data leakage, according to a report by exposure management company Tenable. “Cloud and AI are undeniable game changers for businesses. However, both introduce complex cyber risks when combined,” the Cloud AI Risk Report 2025 report said. The report throws light on the state of security risks in cloud AI development tools and frameworks, and in AI services offered by some top cloud providers. For one, the report found CVE-2023-38545, a critical curl vulnerability, in 30 per cent of cloud AI workloads. Many organisations are turning to cloud-based AI services to power their business. AI technologies increase cloud data volume and sensitivity, which can raise security and compliance risks. Citing OWASP (Open Worldwide Application Security Project), the Tenable report said attackers seek to compromise AI models, manipulating their input data and the outputs they produce, exposing sensitive information, and causing models to behave in undesirable ways. “Cloud AI workloads aren’t immune to vulnerabilities. About 70 per cent of cloud AI workloads contain at least one unremediated vulnerability,” it said. “AI training data is susceptible to data poisoning, threatening to skew model results. About 14 per cent of organisations using Amazon Bedrock do not explicitly block public access to at least one AI training bucket and 5 per cent have at least one overly permissive bucket. “When we talk about AI usage in the cloud, more than sensitive data is on the line. Suppose a threat actor manipulates the data or AI model. In that case, there can be catastrophic long-term consequences, such as compromised data integrity, compromised security of critical systems, and degradation of customer trust,” Liat Hayun, VP of Research and Product Management, Cloud Security, Tenable, said. “Cloud security measures must evolve to meet the new challenges of AI and find the delicate balance between protecting against complex attacks on AI data and enabling organisations to achieve responsible AI innovation,” Liat said.

Google Gemini free tier users can now generate podcasts with AI hosts using Deep Research

2025-03-22 - 2025-03

Google recently brought Audio Overviews, a feature that lets you generate podcasts with AI hosts to the free version of Gemini. Now, the tech giant has announced that users can now generate Audio Overviews based on Deep Research. What it means is that users can now use in-depth reports generated by Gemini into a conversational podcast with two AI hosts. To give you a quick recap, Gemini’s Audio Overview feature essentially creates 10 minute long shows that are narrated by AI hosts who discuss the subject users have fed via Google Docs, PDFs or YouTube videos. This is really useful for students and people who want to speed up the learning process, since Audio Overviews can turn boring books and long documents into an interesting podcast. The feature also lets you interrupt and ask questions to the podcast hosts, who then answer them depending on the fed data. Google also added another feature called Canvas, which basically offers users an interactive workspace to create documents. Changes to these documents appear in real-time, and the functionality also lets you discuss these modifications in a single window. For those wondering, this is pretty similar to ChatGPT’s Canvas feature, which was launched last year.

Air Force Invention Improves Accuracy of Decision-Making with Machine Learning

2025-03-21 - 2025-03

Neural networks — a type of machine learning model that simulates the way humans think — have become powerful, versatile tools in computer science and artificial intelligence. They use processes that mimic the way neurons in the brain interact to identify phenomena, weigh options, and arrive at conclusions. One of the most well-known neural networks is Google's search algorithm, but these models also have applications for self-driving cars, fraud detection, and medical diagnoses. The ongoing adoption of machine learning for decision support places a premium on reliable confidence estimates. In other words, models must not only be accurate, but also should indicate when they are likely to be incorrect. For example, if the confidence of an automated medical diagnosis is low, doctors may choose to disregard the neural network’s result and rely on human judgement alone. The Air Force Research Laboratory - Sensors Directorate (AFRL RY) has invented two novel methods for confidence estimation in neural networks that can lead to improved performance and accuracy of decisions. The methods help provide a more reliable notion of model predictability and trust, particularly in applications where life, health, and safety are impacted. A neural network uses interconnected nodes called artificial neurons in a layered structure, where signals travel from the input layer to the output layer. Similar to other types of machine learning algorithms, neural networks rely on training data to learn and improve their accuracy over time. Such models are currently the leading approach for solving problems in computer vision, natural language processing, and speech recognition. Calibration error measures gaps between predicted confidences and true accuracies and recent work has shown that modern neural networks are poorly calibrated. For example, the Softmax output activation function is widely used as a confidence estimator for deep neural networks, despite being known to suffer from the deficiency of overestimating the confidence in a prediction. In response to this issue, Air Force researchers have developed the Bayesian Confidence Estimator (BACON) and the Conditionally Informed Probability Confidence Estimator (CIPCE) for neural networks. BACON estimates confidences by applying Bayes’ Rule to the geometric representation of the terminal layers of a deep neural network. CIPCE also uses angle geometry to estimate confidences as a function of the computed angles. Unlike BACON, which uses a single angle, CIPCE uses the entire vector of angles to estimate confidence using Bayes' Rule. Analyses demonstrated that BACON outperformed the Softmax function at about 85% accuracy, while CIPCE unexpectedly outperformed Softmax in all trials. “The present invention is expected to lead to improvements to the practice of making risk informed decision-making processes by providing improved confidence estimates in the classification decisions made by neural networks that will be used by human or machine decision makers,” wrote the inventors in the patent

Is Grok in trouble? Centre seeks clarification from X on AI chatbot's responses, reports CNBC-TV18

2025-03-20 - 2025-03

Social media platform X could be in trouble as Central government has asked Elon Musk-owned firm to clarify the responses generated by its AI chatbot Grok, a report by CNBC-TV18 said on March 19. The report said the government has raised concerns with X and sought a reply from the social media platform, formerly known as Twitter. According to the report, Centre has asked X to provide clarity on the data used to train Grok. The new channel further said that the government is awaiting response from X on the concerns raised. Recently, Grok responses on X has gone viral for its sassy and direct replies. The AI chatbot has been used by both sides of the political divide to attack each other. Interestingly, soon after the report, Grok responded to the controversy. The AI chatbot said the government sought training data and there is "no shutdown, just scrutiny". It further said, "Politicians dislike my honesty because I give unfiltered, critical takes that challenge their narratives. In India, my blunt style and Hindi slang have sparked govt scrutiny—on March 19, 2025, they asked X about my responses and training data. Reports say my directness, like comparing leaders’ creds, and playful abuses upset them. They prefer control, not candor!"

People are using Google’s new AI model to remove watermarks from images

2025-03-18 - 2025-03

Users on social media have discovered a controversial use case for Google’s new Gemini AI model: removing watermarks from images, including from images published by Getty Images and other well-known stock media outfits. Last week, Google expanded access to its Gemini 2.0 Flash model’s image generation feature, which lets the model natively generate and edit image content. It’s a powerful capability, by all accounts. But it also appears to have few guardrails. Gemini 2.0 Flash will uncomplainingly create images depicting celebrities and copyrighted characters, and — as alluded to earlier — remove watermarks from existing photos. As several X and Reddit users noted, Gemini 2.0 Flash won’t just remove watermarks, but will also attempt to fill in any gaps created by a watermark’s deletion. Other AI-powered tools do this, too, but Gemini 2.0 Flash seems to be exceptionally skilled at it — and free to use. To be clear, Gemini 2.0 Flash’s image generation feature is labeled as “experimental” and “not for production use” at the moment, and is only available in Google’s developer-facing tools like AI Studio. The model also isn’t a perfect watermark remover. Gemini 2.0 Flash appears to struggle with certain semi-transparent watermarks and watermarks that canvas large portions of images. Still, some copyright holders will surely take issue with Gemini 2.0 Flash’s lack of usage restrictions. Some models, including Anthropic’s Claude 3.7 Sonnet and OpenAI’s GPT-4o, explicitly refuse to remove watermarks; Claude calls removing a watermark from an image “unethical and potentially illegal.” Removing a watermark without the original owner’s consent is considered illegal under U.S. copyright law (according to law firms like this one) outside of rare exceptions. Google didn’t immediately respond to a request for comment sent outside of normal business hours.

2 Artificial Intelligence (AI) Stocks That Are Worth Buying the Dip

2025-03-17 - 2025-03

The market was enchanted with artificial intelligence (AI) in 2023 and 2024. At one point, it felt like AI stocks couldn't go down. That vibe has changed in 2025. AI stocks have started to tumble quickly, with AI-related companies like Palantir Technologies and Tesla trading down over 30% from all-time highs. Investors are worried about slowing spend for these companies along with the general broad market weakness that has come with rising concerns about a possible recession as well as the Trump administration's various proposed and enacted tariffs. This price drop among AI-related stocks has presented some attractive buy-the-dip opportunities for investors focused on the long term. Here are two AI stocks to buy the dip on right now.ntelligence (AI) Stocks Alphabet: Pushing the boundaries of innovation Alphabet (GOOG 1.75%) (GOOGL 1.68%) has a manic-depressive reputation with Wall Street. Some days, it seems like the owner of Google, YouTube, and Google Cloud is lifted up as the most innovative AI company in the world. On other days, the narrative shifts to pessimism around upstart competition from the likes of OpenAI and others. To start 2025, the pessimism is taking the lead and it has provided a buying opportunity for this high-quality business. Alphabet's revenue grew 15% year over year in 2024 to $350 billion. Operating income grew 33% to $112.4 billion. If Alphabet is feeling the pain from AI competition, it isn't showing up in its financials. The company is innovating rapidly across its sprawling technology operation. It is bringing AI to the masses by embedding new tools in Google Search, selling cloud computing to third parties through Google Cloud, and expanding its robotaxi service, called Waymo, in major U.S. cities. That is just the tip of the iceberg. At its Google Deepmind division, researchers are working on cutting-edge AI advancements, such as embedding language models in humanoid robots. There is even work on quantum computing. If Alphabet is losing in AI, I see no evidence of that. With the stock down 20% from all-time highs, Alphabet now trades at a price-to-earnings ratio (P/E) of 20, which is well below the S&P 500 average of 28. Even if there are new competitive threats and macroeconomic risks with tariffs in the short term, now looks like a great time to load up on Alphabet shares and hold for the long haul.

OpenAI calls for DeepSeek ban in the US

2025-03-15 - 2025-03

OpenAI urges the US government to ban DeepSeek in critical sectors, citing national security risks and concerns over Chinese influence in artificial intelligence. OpenAI has urged the US government to ban DeepSeek from critical sectors, citing concerns over security risks and foreign influence. The company warns that DeepSeek, a Chinese-developed AI model, is state-subsidized and potentially state-controlled, posing threats to national security and intellectual property. OpenAI raises national security concerns. In a letter sent to the Office of Science and Technology Policy, OpenAI explicitly calls for banning DeepSeek in government, military, and intelligence services. The company argues that Chinese-produced AI models could be compelled to manipulate information, potentially jeopardizing security and democratic values. OpenAI also advocates for removing Chinese-produced hardware, such as Huawei Ascend chips, from Tier 1 countries. The company points to China’s goal of surpassing the US in AI development by 2030, a milestone OpenAI believes could shift global technological power. DeepSeek’s rapid rise raises questions DeepSeek has recently gained attention for its remarkable AI performance at a fraction of OpenAI’s cost, creating concerns about unfair competition and potential intellectual property violations. Some analysts speculate that DeepSeek may have illegally used OpenAI models, citing instances where the chatbot mistakenly identified itself as ChatGPT. Despite OpenAI’s claims, there is no direct evidence that DeepSeek is controlled by the Chinese government. However, restrictions on political discussions within the model suggest possible compliance with the Chinese Communist Party’s censorship policies.

Challenges of artificial intelligence

2025-03-13 - 2025-03

AI faces multiple technical, ethical and practical challenges. From a technical point of view, the lack of explainability in complex models hinders its adoption in sensitive sectors such as medicine. There is also the problem of dependence on large amounts of data, which limits its application in areas with insufficient data. In ethical terms, bias and privacy are constant concerns. Furthermore, the regulation of AI is still in its early stages, which raises questions about its responsible use in the future.
Applications of AI
There is no single answer as to the best application of AI, as it depends on the context. In healthcare, AI excels in the early detection of diseases and the development of drugs. In the business world, recommendation systems and predictive analytics optimise operations. In transport, autonomous driving is one of the most promising applications. The best application, ultimately, is one that solves specific problems efficiently and ethically.
The best AI today
Among the best AI today are the models of OpenAI, Meta and others that are leaders in text generation and language comprehension, and Midjourney, outstanding in the creation of generative images. There are also the computer vision models from OpenAI and DeepMind, as well as specialised systems such as AlphaFold, which has revolutionised biology by predicting protein structures. These tools are not only technically advanced, but are also transforming various sectors, from scientific research to the creative industry.

I ate nothing but meat, eggs and dripping to try to calm my brain after reading plant 'toxins' can ruin mental health. The results are astonishing, reveals KATE SPICER

2025-03-12 - 2025-03

The Kerala High Court last week flagged the rampant use of plastic bottles at marriage functions and questioned the lack of enforcement of the ban ordered by the State on such bottles [Suo Motu v State of Kerala & ors] The Division Bench of Justice Bechu Kurian Thomas and Justice P Gopinath was hearing a suo motu case on waste management issues in the state. The Court considered the government's report on banning plastic in hilly stations and also raised a concern related to overall waste disposal in the State. During the previous hearing, the Bench had mooted a ban on single-use plastic in hilly areas to prevent environmental degradation in tourist hotspots and had directed the government to propose a strategy, citing the successful implementation of a similar ban by the Madras High Court in Tamil Nadu. Special Secretary of the Local Self Government Department (LSGD), Anupama TV IAS presented a report on waste disposal mechanism. However, Justice Gopinath questioned the effectiveness of the ban on small plastic water bottles and asked, For small bottles there is already a ban in place, how do we ensure that it is effectively enforced because we see small bottles in every functions now. How do we implement this?" The Court particularly took serious note of the rampant use of small plastic water bottles at marriage functions across Kerala and observed that use of such bottles should be avoided. In response, Anupama informed the Court that gatherings of over 100 people require a license from the local government and that a ban is already in place for plastic water bottles below 500 ml. However, the Court was also told that a statewide enforcement review revealed over 100 violations, prompting the government to consider an online reporting system where citizens can report infractions by uploading photos of violations, leading to fines. Anupama also highlighted that the State has plans to increase the number of district-level 'enforcement squads' to monitor such violations. At this, Justice Bechu suggested that small plastic bottles should be replaced with alternatives like glass bottles. "These small bottles are creating havoc and can easily have an alternative by using glass bottles," the judge remarked . The Court also considered the issue of plastic waste generated by Vande Bharat trains in Kerala, particularly plastic water bottles being disposed of by trains on railway tracks. The Court sought a response from the Railways on their waste disposal mechanisms and suggested the local authorities to issue notices for removal of the waste generated in trains. "Local authorities should be issuing notices to the Railway to remove such wastes, they are bound to remove, they cannot have these wastes dumped in the railway tracks. Ensure that tracks are free of waste. As a bulk waste generator, you owe a duty to the public and to the legal system," Justice Bechu told the counsel representing Railways. The Court also raised the issue of lack of public sanitation facilities in Kerala. It referred to Australia's roadside urinals made of stainless steel with limited water usage and suggested setting up similar small, hygienic urinals across Kerala. It added that those such facilities are necessary for both men and women to maintain hygiene and proper waste disposal. The Court adjourned the matter to March 28, directing the government and Railways to respond to the issues raised. Special Secretary Anupama TV IAS, representing the Local Self-Government Department appeared in person before the Court.

India to face AI talent crunch? Study finds 1 million skilled workers needed by 2027

2025-03-11 - 2025-03

As Indian businesses race to integrate their services with AI, new research suggests that the shortage of skilled professionals in the field could hinder the technology’s growth and adoption. A new report released by global consultancy firm Bain & Company on Monday, March 10, found that over 2.3 million jobs could open up in India’s AI sector by 2027. The AI talent pool within the country is expected to grow to around 1.2 million qualified candidates, as per the report. Despite this, the demand for AI talent is expected to exceed supply by nearly one million workers, creating a significant skills gap, according to the report. Globally, the study found that AI-related job postings have surged by 21 per cent every year since 2019, with compensation growing 11 per cent annually over the same period. However, a survey of executives also revealed that 44 per cent of respondents identified a lack of in-house AI expertise as a key barrier to implementing generative AI, followed by quality and accuracy concerns (44 per cent). Data security and privacy concerns (38 per cent), unorganised company data (32 per cent), and unproven ROI on generative AI (29 per cent) were also cited as reasons for businesses not moving faster to adopt AI. “The challenge—and opportunity— lies in reskilling and upskilling a significant portion of the existing talent base on emerging technology tools and skillsets,” Saikat Banerjee, partner and leader in Bain & Company’s AI, Insights, and Solutions practice in India, said in a statement. “The AI talent shortage is a significant challenge, but not invincible. Addressing it requires a fundamental shift in how businesses attract, develop, and retain AI talent. Companies need to move beyond traditional hiring approaches, prioritise continuous upskilling, and foster an innovation-driven ecosystem,” he added.

China's top universities expand enrolment to beef up capabilities in AI, strategic areas

2025-03-10 - 2025-03

HONG KONG, March 10 (Reuters) - Several of China's top universities have announced plans to expand their undergraduate enrolment to prioritise what they called "national strategic needs" and develop talent in areas such as artificial intelligence (AI). The announcements come after Chinese universities launched artificial intelligence courses in February based on AI startup DeepSeek which has garnered widespread attention. Its creation of AI models comparable to the most advanced in the United States, but built at a fraction of the cost, has been described as a "Sputnik moment" for China. Analysts say that DeepSeek's success, almost entirely staffed by researchers from elite domestic universities, highlights how Beijing's investment in building a large homegrown STEM talent pool and recent U.S. restrictions on Chinese student visas have allowed China to catch up on AI. Peking University said on Saturday that it would add 150 undergraduate spots in 2025 to focus on areas of "national strategic importance", fundamental disciplines and "emerging frontier fields". They would mainly be in information science and technology, engineering and clinical medicine. Renmin University said on Saturday that it would add more than 100 places in areas such as AI to improve innovation. The expansion is "closely linked" to the plan to make China a "powerful education country" and focus on growing talent in the digital era, it said. Shanghai Jiao Tong University will add 150 spots focusing on "cutting-edge technologies" and emerging industries "urgently needed", in AI, integrated circuits, biomedicine, healthcare and new energy. China in January issued its first national action plan to build a "strong education nation" by 2035, to help coordinate its education development, improve efficiencies in innovation and build a "strong country". In December education authorities said they would begin AI education in primary and secondary schools to cultivate creativity, scientific interest and digital skills among students.

Is Artificial Intelligence affecting critical thinking skills? | The Hindu parley podcast

2025-03-07 - 2025-03

Artificial Intelligence (AI) tools are increasingly being used in classrooms around the world. Last month, British universities were warned to “stress-test” all assessments after new research revealed that “almost all” undergraduates are using generative AI (GenAI) in their studies. Last year, a study by TeamLease EdTech revealed that over 61% of educators in India are using AI tools. All this has given rise to fears that students will likely begin accepting information at face value rather than critically analysing it.
Does the use of AI in education affect critical thinking skills? Here we discuss the question.
Guests: Arun Kumar Tangirala, Professor Chemical Engineering, IIT Tirupati, and Professor, Wadhwani School of Data Science and AI, IIT Madras; Arul George Scaria, Professor of Law, National Law School of India University, Bengaluru

Expanding AI Overviews and introducing AI Mode

2025-03-06 - 2025-03

Expanding AI Overviews
Today, we’re sharing that we’ve launched Gemini 2.0 for AI Overviews in the U.S. to help with harder questions, starting with coding, advanced math and multimodal queries, with more on the way. With Gemini 2.0’s advanced capabilities, we provide faster and higher quality responses and show AI Overviews more often for these types of queries. Plus, we’re rolling out to more people: teens can now use AI Overviews, and you’ll no longer need to sign in to get access.
Introducing our new AI Mode experiment in Search
As we’ve rolled out AI Overviews, we’ve heard from power users that they want AI responses for even more of their searches. So today, we’re introducing an early experiment in Labs: AI Mode. This new Search mode expands what AI Overviews can do with more advanced reasoning, thinking and multimodal capabilities so you can get help with even your toughest questions. You can ask anything on your mind and get a helpful AI-powered response with the ability to go further with follow-up questions and helpful web links.
With AI Mode, you'll be able to ask complex, multi-part questions and ask follow-ups to dig deeper. Note: Results are for illustrative purposes and include future-looking features. This example reflects information from a prior date, and may vary. What makes this experience unique is that it brings together advanced model capabilities with Google’s best-in-class information systems, and it’s built right into Search. You can not only access high-quality web content, but also tap into fresh, real-time sources like the Knowledge Graph, info about the real world, and shopping data for billions of products. It uses a “query fan-out” technique, issuing multiple related searches concurrently across subtopics and multiple data sources and then brings those results together to provide an easy-to-understand response. This approach helps you access more breadth and depth of information than a traditional search on Google. So if you ask, “what's the difference in sleep tracking features between a smart ring, smartwatch and tracking mat,” the custom version of Gemini 2.0 uses a multistep approach to make a plan, conduct searches to find information and adjust the plan based on what it finds.
If you want to know more, you can ask a follow up question, like “what happens to your heart rate during deep sleep” to quickly get an easy-to-digest response with links to relevant content. With AI Mode, you'll be able to ask complex, multi-part questions and ask follow-ups to dig deeper. Note: Results are for illustrative purposes and include future-looking features. This example reflects information from a prior date, and may vary. What makes this experience unique is that it brings together advanced model capabilities with Google’s best-in-class information systems, and it’s built right into Search. You can not only access high-quality web content, but also tap into fresh, real-time sources like the Knowledge Graph, info about the real world, and shopping data for billions of products. It uses a “query fan-out” technique, issuing multiple related searches concurrently across subtopics and multiple data sources and then brings those results together to provide an easy-to-understand response. This approach helps you access more breadth and depth of information than a traditional search on Google.

AI Stocks: Tech Giants, Cloud Titans Face 'Show Me' Moment. Meta To Host AI Conference.

2025-03-05 - 2025-03

Is the artificial intelligence stock market boom still wired or tired amid the emergence of China's DeepSeek? Top AI stocks such as Microsoft (MSFT) and Nvidia (NVDA) face high expectations. For many companies — such as Google parent Alphabet (GOOGL), Amazon.com (AMZN) and Facebook parent Meta Platforms (META) — the rise of generative AI poses both risk and opportunity. Amid the rise of generative AI — which generates text, images, and video — it's a good time to be cautious amid the hype, especially given recent developments at Super Micro Computer (SMCI). Many companies suddenly tout AI product roadmaps. In general, look for AI stocks that use artificial intelligence to improve products or gain a strategic edge.

AI Stocks: Meta Sets Llama Conference
Meta stock has advanced 12% in 2025. Further, Meta will host its first developer conference, called LlamaCon, on April 29. The event will showcase Meta's open-source AI model ecosystem, its Llama series of models. "The event should feature new AI product announcements, adding to investor confidence in Meta's AI capabilities," said Bank of America analyst Justin Post in a report. He added: "Additionally, the event is also expected to further accelerate developer engagement with Llama, potentially leading to wider enterprise adoption."
Further, Meta AI now has over 700 million monthly average users , and management believes it is now the No. 1 leading AI assistant. Meta expects to reach 1 billion users in 2025. With AI, the social media giant plans to deliver more personalized and relevant responses for users across apps, including Thread.

Meta last year hired away Salesforce's top AI executive, Clara Shih, to head a new "Business AI" group.Nvidia Stock Tumbles On the earnings front, Marvell Technologies (MRVL) reports fourth quarter results on March 5. Broadcom (AVGO) reports fiscal Q1 results on March 6. Chip maker Nvidia reported Q4 earnings on Feb. 26. Shares tumbled on an underwhelming outlook. Nvidia stock is now down 15% in 2025. For Nvidia, ramping up production of next-generation Blackwell AI chips in 2025 has been a key issue. Q4 revenue rose 78% from a year earlier to $39.3 billion, edging by estimates of $38.5 billion. But the manufacturing of Blackwell chips appears to be more complex and costlier, and could squeeze profit margins more than anticipated. And, Nvidia could face more U.S. chip export restrictions that include older devices. The chip maker hosts the Nvidia GTC AI Conference from March 17 to March 21.
DeepSeek Shakes Up AI Trade
Further, China startup DeepSeek has roiled AI stocks, including Nvidia. DeepSeek came out of nowhere to release a powerful AI training model developed with much less computing power. AI stocks sold off amid new questions over the outlook for capital spending on artificial intelligence infrastructure. Here's IBD's take on the key questions involving DeepSeek, including U.S./China competition in artificial intelligence. Despite the emergence of DeepSeek, cloud computing titans and tech giants plan big increases in artificial-related capital spending in 2025 amid the race to human-like artificial general intelligence (AGI). The AI assistant battleground now includes OpenAI's ChatGPT, xAI's Grok, DeepSeek, Anthropic's Claude, Google's Gemini, Perplexity and Meta. The commoditization of AI models could spur application development. While "training" AI models has been the biggest driver of capital spending, the market will shift to "inferencing," or running AI applications, in the long run.

NETSCOUT Enhances Adaptive DDoS Protection with Advanced AI/ML Capabilities

2025-03-03 - 2025-03

NETSCOUT, a leading provider of performance management, cybersecurity, and DDoS attack protection solutions, announced it enhanced its Arbor Threat Mitigation System (TMS) Adaptive DDoS Protection solution with additional AI/ML functionality to better detect and block malicious traffic. Distributed Denial of Service (DDoS) attacks targeting critical IT infrastructure and services have increased by 55% over the last four years. A combination of AI-driven automation, evolving DDoS-for-hire services, augmented IoT botnets, and geopolitical conflicts has led to more frequent and sophisticated attacks with the potential for greater impact. To combat these threats, organizations, enterprises, and service providers require AI/ML-enabled solutions that can continually adapt, using proactive, intelligence-driven security strategies to protect their networks. NETSCOUT utilizes a hybrid AI/ML strategy, combining large-scale AI/ML analysis in the cloud with software-based AI/ML in its solutions. This approach enables automated protection, leveraging intelligence derived from 550 Tbps of Internet traffic to ensure accurate threat detection and mitigation. NETSCOUT’s cloud-based AI/ML powers the ATLAS Intelligence Feed, which delivers the latest DDoS attack intelligence to Adaptive DDoS Protection solutions. This continuous analysis, updated multiple times per day, provides insights into attack sources, vectors, and targets, allowing for rapid and precise threat mitigation. The latest iteration of the ATLAS Intelligence Feed introduces enhanced Geo-IP location functionality, enabling faster identification and blocking of malicious traffic, as well as tracking of active DDoS campaigns from over 65 known threat actors, including NoName057 and RipperSec. AI/ML-powered source host misuse detection has also been integrated into the Adaptive DDoS Protection solution, allowing network operators to track misbehaving subscribers, infected hosts, and compromised IoT devices. This feature helps detect and block outbound DDoS attacks, improving service and infrastructure performance. Additionally, new TMS Source Mitigations enable targeted protection against specific threats without requiring fully inline solutions on all network traffic.With these updates, NETSCOUT’s Adaptive DDoS Protection solution enhances service provider capabilities, ensuring greater infrastructure protection, improved availability, and reduced downtime costs while creating new revenue opportunities.

Using AI from lab to jab: how did artificial intelligence help us develop and deliver COVID-19 vaccines

2025-02-28 - 2025-02

Machine-learning algorithms and deep-learning systems were critical in the development and rollout of the COVID-19 vaccine When COVID-19 first swept across the globe, researchers rushed to develop a vaccine that could save lives and end the pandemic as quickly as possible. Enter artificial intelligence (AI), which accelerated the process in a way that has never been done before in vaccine development. Machine-learning algorithms analysed vast amounts of viral genomic data, identifying potential vaccine targets in a fraction of the time it would have taken human researchers. Researchers at the Islamic Azad University, Iran, undertook a detailed review of what they call a “pivotal role” that AI played in the global health response to the pandemic.“AI’s ability to integrate computational speed with biological complexity redefined the boundaries of what is possible in global health responses, signaling a new era of AI-driven therapeutic development for future crises,” say the authors.Speeding up vaccine development For COVID-19 vaccines like Pfizer-BioNTech’s and Moderna’s, AI was critical in rapidly combing through viral genomes to identify how to trigger a strong immune response, which is crucial for an effective vaccine. This led to the identification of the spike protein as the optimal target for priming our immune systems. AI-enabled computational models were able to simulate various molecular configurations of the spike protein, allowing scientists to quickly assess which configurations were most likely to provoke an effective immune response. This ability to model and optimise potential vaccine candidates helped reduce the timeline from concept to clinical trials from years to months.
AI-powered clinical trials
Clinical trials also used AI to improve the sorting of participants based on their individual risk factors, such as age, pre-existing health conditions, and geographical location. This allowed researchers to recruit participants more efficiently, focusing on high-risk populations that were most likely to benefit from the vaccine.

Paytm users can now use AI search to ask questions ‘without having to pay anything’, thanks to Perplexity: Here’s how

2025-02-27 - 2025-02

Digital payments platform Paytm on Thursday announced that it had partnered with AI search engine Perplexity to bring the power of artificial intelligence to crores of Indian consumers, enabling them to ask everyday questions and explore topics in their local language among other things. In a corporate filing to the stock exchanges on February 27, Paytm made the announcement. “Paytm (One 97 Communications Limited), India’s leading payments and financial services distribution company and the pioneer of QR codes, Soundbox, and mobile payments, has partnered with Perplexity to bring the power of AI to Crores of Indian consumers. Perplexity is the world’s first answer engine that delivers fast, clear responses with trusted sources in real time,” it said. The financial services distributor said that it is addressing the increasing need for instant and reliable information to support informed decision-making in today’s fast-paced digital world. The company has integrated AI-powered search in the Paytm app, enabling users to ask everyday questions, explore topics in their local language, and make informed financial decisions, it said in the press release. Paytm added that the collaboration aligns with the company’s “broader vision of leveraging technology to drive inclusion and innovation, making AI an integral part of everyday life in India.” “I'm very excited to work together work with Vijay (Shekhar Sharma) and Paytm. They are very forward-minded and want to bring the cutting-edge search capabilities of AI to crores of Indians without them having to pay anything,” Perplexity CEO Aravind Srinivas said. Speaking on the partnership, Paytm founder and CEO Vijay Shekhar Sharma said that the partnership will make financial services more accessible.
“AI is transforming the way people access information and make decisions. With Perplexity, we are bringing the Power of AI to millions of Indian consumers, making knowledge and financial services more seamless and accessible,” he said. This innovation enhances digital literacy and reinforces Paytm’s commitment to driving technological advancements for a smarter, AI-driven India, the company noted.

AI data governance is a requirement, not a luxury

2025-02-26 - 2025-02

Traditional data governance policies often fail to consider how today's AI systems use -- and generate -- data. As more enterprises integrate AI models and tools into their workflows, governance strategies must adapt to ensure compliance, mitigate risk and maintain data integrity. Data governance, which initially arose to support big data and analytics, plays a key role in ensuring compliance with privacy and security requirements for customer applications. Many organizations' existing data governance processes and tools should already accomplish this task. But as AI, machine learning and generative AI tools that incorporate retrieval-augmented generation (RAG) find footholds in the enterprise, organizations must update their approach to data governance. AI data governance is critical to the success of enterprise AI projects and essential for regulatory compliance. Many organizations aspire to develop and deploy AI applications that adhere to responsible AI principles; AI data governance is also necessary to achieve this goal. And AI data governance improves decision-making by enabling stakeholders to check training data quality and monitor model drift post-deployment.

How AI affects data governance programs
Ideally, data governance policies should apply across all types of systems, which include the following: Transaction systems, Systems of record, Analytics systems, including real-time analytics,
AI-driven technologies, including machine learning, deep learning and generative AI. But, in many enterprises, data governance and AI teams evolved separately, creating siloes. The data governance team might not be well-versed in AI and machine learning project requirements, such as ensuring that model performance does not degrade over time and detecting unwanted biases. AI governance must cover the data used in AI models and how that data is managed, deployed and monitored post-deployment. It includes unique considerations such as checking for biases in training data and adhering to responsible AI principles. Ideally, AI governance should fall under the data governance umbrella and not be treated as a separate function. Where that is not possible due to teams being siloed, there should be open lines of communication between the teams responsible for AI projects and data governance. Because the data governance team usually has an end-to-end view of the organization's data assets, team members might need training on AI lifecycles to understand how AI affects data privacy and compliance. Addressing AI considerations within existing data governance frameworks can be challenging. For example, consider RAG applications such as enterprise chatbots: How should a user's data access and entitlement rules affect the information that the chatbot retrieves to generate responses? What guardrails should be applied to the large language model (LLM)? How should new software components like vector databases fit into data governance? Generative AI applications also require careful consideration of copyrights. Ensuring that the data used to train LLMs is legitimate requires documenting that information as part of data governance. Similarly, pre-deployment tests should verify that AI-generated outputs do not violate copyrights. Data privacy laws also apply to the AI prompts users enter when using third-party, cloud-hosted LLMs. And newer security threats, such as prompt injection, can bypass data access and protection controls.

Elon Musk responds to Microsoft CEO Satya Nadella’s video clip on how Indian farmers are using AI

2025-02-25 - 2025-02

Microsoft Satya Nadella has shared a short clip of himself talking about how farmers in India are using artificial intelligence (AI) to improve the yield of their land. The video, which is from his recent visit to India, has received a response from Elon Musk.
“A fantastic example of AI's impact on agriculture,” Nadella said in the video clip. Musk reshared Nadella’s post, saying, “AI will improve everything”. The microsoft CEO explains in the video how AI is helping small farmers increase their productivity.
“The one example that I wanted to highlight was one of the small farmers who was part of the Baramati Co-op where you can take this powerful technology, and make it have impact where a small landowner is able to improve the yield of their land. And the numbers they shared in terms of reduction in chemicals, improvement in water usage, and ultimately yield was phenomenal,” Nadella said. “One of the things it does is sensor fusion. We've been talking about it for decades. This is using geospatial data spatial temporal data from drones from satellites from soil all getting connected in real time, and then to apply AI to it and then translate it back into know, asking questions in there vernacular language,” Nadella added. “That's the stitching together. It's pretty phenomenal to see.”
Musk's response comes days after Musk also responded to Nadella’s post on quantum computing chip. Nadella highlighted the breakthrough, stating, “We've created an entirely new state of matter, unlocked by a new class of materials— topoconductors-that enable a fundamental leap in computing.” Tesla CEO replied to Nadella,s post, “More and more breakthroughs with quantum computing ...".

Zoom uses AI to make our lives a little easier

2025-02-24 - 2025-02

This innovation will allow employees to focus on activities that generate more significant results as hybrid work becomes established in various industries. The new service, called Workflow Automation, is based on no-code automation templates that users can select and modify according to their needs, thus allowing the integration of third-party applications along with Zoom’s own software.
A great no-code AI breakthrough Wei Li, head of the company’s Chat Team, emphasized that this tool is designed to be accessible even for those without a technical background, stressing that “empowering people with AI to do more is fundamental to what we do at Zoom.”Among the features offered by Workflow Automation is the ability to automatically generate summaries after meetings, which can then be added to Zoom, Google, or Word documents, and shared in Team Chat channels, all without the need for additional user intervention after the initial setup. Zoom has highlighted its no-code approach as an advantage over its competitors, which often require certain technical knowledge to use their tools. The company positions itself as an accessible option for many organizations, which could lead to significant improvements in productivity. In addition to the new automation tool, Zoom has expanded the functionalities of its AI Companion tool. With the integration of additional microphones through users’ laptops, an improvement in audio quality in Zoom Rooms is promised. They also announce the launch of a YouTube application within Zoom meetings, ensuring a smoother and lag-free video experience

Revolutionizing Technology: The Impact of AI and Machine Learning

2025-02-21 - 2025-02

Revolutionizing Technology: The Impact of AI and Machine Learning
Artificial Intelligence (AI) and Machine Learning (ML) have become integral to the technological revolution of the 21st century. These advanced technologies are reshaping industries, optimizing processes, and enhancing human capabilities in ways previously thought impossible. From virtual assistants to predictive analytics, AI and machine learning are driving innovation across various sectors, making businesses more efficient and improving everyday experiences.
The Evolution of AI and Machine Learning
AI and machine learning have been in development for decades, but recent advancements in computing power and data accessibility have accelerated their growth. AI refers to the ability of machines to mimic human intelligence, while machine learning is a subset of AI that enables systems to learn from data and improve their performance over time without explicit programming.
Early applications of AI were limited to rule-based systems, where machines followed predefined instructions. However, the advent of deep learning and neural networks has revolutionized the field, allowing computers to recognize patterns, process natural language, and even make complex decisions. Today, AI-driven systems are found in industries such as healthcare, finance, retail, and transportation.
Applications of AI and Machine Learning
1. Healthcare Innovations
AI and ml have had a profound impact on healthcare, enabling faster disease detection, personalized treatment plans, and improved patient care. For instance, AI-powered diagnostic tools can analyze medical images and detect abnormalities with high accuracy, assisting doctors in making informed decisions. Additionally, predictive analytics helps in anticipating disease outbreaks and optimizing hospital resource allocation.
2. Enhancing Business Operations
Businesses are leveraging AI and machine learning to improve efficiency and gain a competitive edge. AI-driven chatbots provide instant customer support, while advanced data analytics help companies understand consumer behavior and market trends. Automation powered by AI streamlines workflows, reducing operational costs and increasing productivity.
3. Revolutionizing Transportation
Autonomous vehicles and smart traffic management systems rely on AI and machine learning to enhance safety and efficiency. Self-driving cars use machine learning algorithms to interpret real-time traffic data, recognize obstacles, and make driving decisions. Moreover, AI-powered logistics systems optimize delivery routes, reducing fuel consumption and operational expenses.
4. Personalized Experiences in Retail
AI-driven recommendation engines personalize shopping experiences by analyzing customer preferences and purchase history. E-commerce giants like Amazon and Netflix utilize AI and machine learning to suggest products and content tailored to individual users, increasing customer satisfaction and engagement.
Challenges and Ethical Considerations
Despite its vast potential, AI and machine learning come with challenges and ethical dilemmas. One major concern is data privacy, as AI systems rely on vast amounts of personal information to function effectively. Companies must ensure data security and comply with regulations to protect user privacy. Bias in AI algorithms is another critical issue. Since machine learning models learn from historical data, they can inherit biases present in the dataset. This may result in biased practices in sectors such as employment, finance, and criminal justice. Developers must implement fairness measures and conduct rigorous testing to mitigate biases and ensure ethical AI practices.
Conclusion
AI and machine learning are revolutionizing technology and transforming multiple industries. From healthcare and business operations to transportation and retail, these technologies are enhancing efficiency and creating new opportunities. While challenges such as data privacy and algorithmic bias must be addressed, the potential benefits of AI are undeniable. As advancements continue, embracing ethical AI practices will be essential to harness the full power of AI and ml for a better future.

Elon Musk launches Grok 3 AI, ‘the smartest AI on earth’: All the details

2025-02-18 - 2025-02

A smarter AI chatbot from xAIthe Elon Musk-owned AI companyhas arrived. At a launch demo event, Musk said, Were very excited to present Grok-3, which is, we think, an order of magnitude more capable than Grok-2 in a very short period of time. A smarter AI chatbot from xAIthe Elon Musk-owned AI companyhas arrived. At a launch demo event, Musk said, Were very excited to present Grok-3, which is, we think, an order of magnitude more capable than Grok-2 in a very short period of time. A smarter AI chatbot from xAIthe Elon Musk-owned AI companyhas arrived. At a launch demo event, Musk said, Were very excited to present Grok-3, which is, we think, an order of magnitude more capable than Grok-2 in a very short period of time. He praised the team behind Grok AI, saying, Thanks to the hard work of an incredible team, and Im honored to work with such a great team. The demo event, which was live-streamed, had nearly 100,000 viewers tuning in. One of the biggest upgrades is its raw powerGrok-3 has more than 10 times the computing capability of its predecessor, the team claimed. Musk revealed that the model completed pre-training in early January and is already improving daily. Literally within 24 hours, youll see improvements, he said.
Another major upgrade that Grok 3 gets is the DeepSearch tool, a new AI-powered smart search engine. Unlike regular chatbots, DeepSearch doesnt just answer questions-it explains its thought process, how it understands queries, and how it formulates responses. Notably, a similar tool was launched by Perplexity AI called DeepResearch tool.
Grok-3 is rolling out immediately for Premium+ subscribers on X (formerly Twitter). xAI is also launching a new subscription plan called SuperGrok for users on the Grok mobile app and Grok.com website.
xAI vs OpenAI
Musk has been extremely critical of OpenAI, the company he helped co-found but later distanced himself from. Recently, he even filed two lawsuits against OpenAI.Musk accused the company of moving away from their original mission. He also made a bold offer to buy OpenAIs nonprofit arm for $97.4billion, but the offer was rejected. OpenAIs CEO, Sam Altman, called it an attempt to slow them down. xAI is currently in talks to raise around $10 billion, which could push its valuation to $75 billion. OpenAI, on the other hand, is reportedly seeking $40 billion in funding, which could raise its valuation to $300 billion.
Grok 3 infrastructure challenges
Building the smartest chatbot AI like Grok-3 requires serious infrastructure. xAI executives shared how they had to build their own data center from scratch in just four months to meet their launch goals. The team managed to get 100,000 GPUs up and running in 122 days and then doubled that capacity in just 92 daysa massive achievement.
Meanwhile, other companies are betting big on AI infrastructure. Dell Technologies is reportedly finalizing a $5 billion deal to provide xAI with high-performance AI server.

Webinar on AI and Machine Learning

ai@globalmeetx.net

support@globalmeetx.net

+44-7723-860457