Software Spotlight: CrowdStrike CrowdStrike Falcon® Cloud Security is a unified cloud security platform that protects infrastructure, applications, data, AI, and SaaS across hybrid and multi-cloud environments. It enables organizations to consolidate tools, reduce complexity, and stop breaches. Code-to-Cloud Protection: Streamlines security with a single agent and agentless architecture that unifies cloud point products and eliminates security gaps. Built to Stop Breaches: Delivers advanced runtime protection, managed threat hunting, and native Cloud Detection and Response (CDR) to stop breaches in real time. Hybrid Cloud Security: Provides unified visibility and security across cloud and on-premises environments for seamless protection in hybrid architectures. Cloud security posture management, or CSPM tools, are automated security solutions designed to continuously monitor and assess cloud infrastructures, services, and applications for misconfigurations and compliance issues. These tools are more important than ever as more organizations now leverage the multicloud approach to cloud adoption, a practice that comes with configuration and security compliance complexities. According to a Fortinet-sponsored report conducted by cybersecurity experts, many organizations are now increasingly wary of AI-based threats, prompting them to have heightened concern about cloud security. To address this challenge, many cloud-first organizations now deploy CSPM tools to help them monitor, identify, alert, and remediate compliance risks and misconfigurations in their cloud environments. To determine which CSPM tool is best suited for your organization, I’ve compiled a list of the top CSPM solutions for 2024. Semperis Employees per Company Size Micro (0-49), Small (50-249), Medium (250-999), Large (1,000-4,999), Enterprise (5,000+) Large (1,000-4,999 Employees), Enterprise (5,000+ Employees) Large, Enterprise Features Advanced Attacks Detection, Advanced Automation, Anywhere Recovery, and more ESET PROTECT Advanced Employees per Company Size Micro (0-49), Small (50-249), Medium (250-999), Large (1,000-4,999), Enterprise (5,000+) Any Company Size Any Company Size Features Advanced Threat Defense, Full Disk Encryption , Modern Endpoint Protection, and more ManageEngine Log360 Employees per Company Size Micro (0-49), Small (50-249), Medium (250-999), Large (1,000-4,999), Enterprise (5,000+) Micro (0-49 Employees), Small (50-249 Employees), Medium (250-999 Employees), Large (1,000-4,999 Employees), Enterprise (5,000+ Employees) Micro, Small, Medium, Large, Enterprise Features Activity Monitoring, Blacklisting, Dashboard, and more What is cloud security posture management? CSPM tools can help users maintain a secure cloud posture by recommending best practices and enforcing security policies across all cloud accounts and services. These policies can include access controls, encryption settings, network configurations, and more. By automating the enforcement process, CSPM software minimizes the risk of misconfigurations and helps defend against threats or outside attacks. SEE: Brute Force and Dictionary Attacks: A Guide for IT Leaders (TechRepublic Premium) Best cloud security posture management software comparison The table below provides a comparison of key features available in each CSPM option. Best cloud security posture management software Here is a rundown of the seven best CSPM software choices in 2024, highlighting their features, pricing plans, pros, and cons. Orca Security: Best for cloud workloads Image: Orca Orca presents users with a CSPM tool that scans their workloads and maps the results into a centralized platform. It can analyze risks and identify situations where seemingly unrelated issues could lead to harmful attack paths. With these insights, Orca prioritizes risks, minimizing the burden of excessive alerts for users. SEE: Everything You Need to Know about the Malvertising Cybersecurity Threat (TechRepublic Premium) Orca also facilitates continuous monitoring for cloud attacks. It features a visual graph to give insight into an organization’s potential attack surface and the attacker’s end target within a cloud environment. Regarding compliance, Orca provides compliance features that enable cloud resources to adhere to regulatory frameworks and industry benchmarks, including data privacy requirements. The platform unifies compliance monitoring for cloud infrastructure workloads, containers, identities, data, and more, all within a single dashboard. Why I chose Orca Security I have Orca Security on this list because it’s a quality solution for organizations that primarily work on the cloud. Its risk analysis and identification of cloud workloads make it a useful tool to combat unnoticed threats. In my view, its extensive reporting and insights functionality, covering an organization’s attack surface, is another feature inclusion that makes it a top choice for those looking to address vulnerabilities or prevent future attacks. Pricing Orca offers a 30-day free trial. Contact Orca to get a quote. Features Cloud compliance. Unified Data Model. Continuous monitoring. Orca Security Score. Attack path analysis. PII detection. Malware detection. The Orca misconfiguration alert dashboard. Image: Orca Orca Security pros and cons Pros Cons Users can create personalized views of Orca’s Risk Dashboard. No pricing information is available on its website. This solution offers a 30-day free trial. It helps organizations meet compliance with PCI-DSS, GDPR, HIPAA, and CCPA. Users can generate comprehensive cloud security reports and share them across various channels. Users can write their own alert queries or use over 1,300 prebuilt system queries. Prisma Cloud: Best for multicloud environments Image: Prisma Prisma Cloud by Palo Alto Networks offers comprehensive visibility and control over the security posture of deployed resources in multicloud environments. The solution can help users implement instant configurations with over 700 pre-defined policies from more than 120 cloud services. That feature can aid organizations in correcting typical multicloud misconfigurations, preventing potential security breaches, and developing custom security policies. With Prisma Cloud, users can also benefit from continuous compliance posture monitoring and one-click reporting, offering coverage for various regulations and standards, including CIS, GDPR, HIPAA, ISO-27001, NIST-800, PCI-DSS, and SOC 2. The solution also provides custom reporting. SEE: What is Cloud Security? Fundamental Guide (TechRepublic) Prisma Cloud offers network threat detection and user entity behavior analytics features, allowing customers to identify unusual network activities, DNS-based threats, and insider threats by monitoring billions of flow logs received every week. Why I chose Prisma Cloud I selected Prisma Cloud due to it being a high quality option for organizations already employing multicloud environments. With more and more companies adopting multicloud, I particularly appreciate Prima Cloud’s pre-defined policies and built-in network threat detection. These can help catch holes or risk areas across multiple cloud providers. This is especially crucial with multicloud environments, as different

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Shutterstock is reshaping how AI companies access training data through a novel “research license” approach, launching first with AI creative technology company Lightricks. The partnership, announced today, allows Lightricks to train its open-source video generation model LTXV using Shutterstock’s extensive HD and 4K video library. The new licensing model addresses a critical challenge in AI development: the high cost of accessing quality training data. It enables companies to start with a smaller research license for testing and experimentation before committing to more expensive commercial licenses. Making ethical AI development more accessible for startups “Many companies and model trainers have taken the route of unauthorized data scraping [instead of] making the necessary investment to achieve the quality and level of trust needed to develop commercially viable models,” said Daniel Mandell, Shutterstock’s global head of data licensing & AI, in an exclusive interview with VentureBeat. “However, we don’t think that financial investment should be a barrier for those looking to enter this space with an ethical approach.” This two-phase approach could transform how startups approach AI development. Craig Andrews, Lightricks’ global PR manager, describes it as “a turning point for smaller, more agile developers who want to explore innovative applications of generative AI without the heavy upfront costs of traditional licensing.” Legal protection and fair compensation in the age of AI The timing is significant, coming amid increasing legal scrutiny of AI training data practices. Several major AI companies face lawsuits over alleged unauthorized use of copyrighted material for model training. Shutterstock’s approach offers a legitimate alternative while ensuring content creators receive compensation. “We’re setting a standard for ethical AI development while ensuring that creators are fairly compensated for their work,” Andrews explains. “This approach not only fosters trust in the creative ecosystem but also establishes a sustainable framework for responsible AI innovation.” Revenue sharing: A win-win for creators and AI companies Shutterstock has implemented a revenue-sharing model where contributors receive 20% of the revenue from data licensing deals. Contributors can also opt out of having their content used for AI training, though Mandell notes only about 1% have chosen to do so. Lightricks plans to use the licensed video data to enhance LTXV, its open-source video generation model released last month. The model has already gained significant traction, with “thousands of downloads on GitHub and Hugging Face,” according to Andrews. One notable use case is real-time video generation for interactive ecommerce. The partnership aims to address technical challenges in AI video generation, particularly motion consistency in longer videos. “One of the biggest technical hurdles in AI video generation is achieving consistent motion and structure over longer video segments without sacrificing quality,” Andrews says. “Shutterstock’s high-quality video library provides an extensive dataset that helps us address this challenge.” For Shutterstock, this partnership represents a strategic shift in its business model. The company has already established partnerships with major AI companies including Nvidia, Meta, and OpenAI. Mandell emphasizes that the research license model could democratize access to high-quality training data for smaller organizations and research institutions. Setting new industry standards for ethical AI development The collaboration also reflects a growing trend toward transparency and ethical considerations in AI development. Lightricks made LTXV open-source to promote collaboration and innovation, while Shutterstock’s licensing approach ensures proper compensation for content creators. “The important message here is that companies, no matter the size or funding, no longer have an excuse to scrape unlicensed content for training purposes,” Mandell concludes. “There is a better way to enter this evolving market.” This partnership could set a new standard for how AI companies access training data, potentially influencing industry practices as concerns about the sources of AI training data continue to grow. The success of this model could determine whether other content providers follow Shutterstock’s lead in creating more flexible, accessible licensing options for AI development. source

安記海味董事總經理潘權輝表示，即使目前香港經濟狀況一般，他仍對工展會有信心，更表示有顧客半年前開始查詢工展會詳情。今年安記海味採取薄利多銷的策略，例如兩盒一人份佛跳牆只售130元、100元3份冬菇等，希望成功令生意額按年增一成。 工展會今日開鑼，廠商會會長盧金榮表示，與中港巴士公司及旅行社合作，向內地旅客派發免費門券，相信「一簽多行」帶動下，內地旅客入場人次將增加。大會亦會向每日首180名入場人士，派發價值500元的免費福袋。 廠商會會長盧金榮。 多個商戶相信生意按年增長至少一成，紛紛推出優惠吸客，包括1元鮑魚海味福袋。有商戶引述港人反映北上消費太多，期待到工展會「掃貨」。 LinkedIn Email Facebook Twitter WhatsApp source

The Federal Trade Commission (FTC) on Tuesday announced action against Gravy Analytics and Venntel Inc. and a separate action against Mobilewalla that would ban the companies from selling sensitive location data. The FTC’s complaint against the companies alleges Virginia-based Gravy Analytics and its subsidiary Venntel violated the FTC Act by unfairly selling sensitive consumer location data, and by collecting and using consumers’ location data without consent for commercial and government uses. Gravy Analytics, the complaint says, also sold health and medical decisions, political activities, and religious views collected from location data. In the case of Georgia-based Mobilewalla, the FTC alleges the company collected more than 500 million unique consumer advertising identifiers paired with precise location data between January 2018 and June 2020. The company sold the raw data to third parties, including advertisers, data brokers, and analytics firms, the FTC says. In a statement, FTC Chair Lina Khan said, “Persistent tracking by data brokers can put millions of Americans at risk, exposing the precise locations where service members are stationed or which medical treatments someone is seeking. Mobilewalla exploited vulnerabilities in digital ad markets to harvest this data at a stunning scale.” In a message to InformationWeek, Mobilewalla CEO Anindya Datta pushed back on the FTC’s case, but accepted the results. “Mobilewalla respects consumer privacy and has been evolving our privacy protections throughout our history as a company,” he says. “While we disagree with many of the FTC’s allegations and implications that Mobilewalla tracks and targets individuals based on sensitive categories, we are satisfied that the resolution will allow us to continue providing valuable insights to businesses in a manner that respects and protects consumer privacy.” FTC had strong words for the companies’ practices. “Surreptitious surveillance by data brokers undermines our civil liberties and puts servicemembers, union workers, religious minorities, and others at risk,” Samuel Levine, director of the FTC’s Bureau of Consumer Protection, said in a statement. “This is the FTC’s fourth action taken this year challenging the sale of sensitive location data, and it’s past time for the industry to get serious about protecting Americans’ privacy.” The FTC also alleged Gravy Analytics and Venntel obtained consumer location information from other data suppliers and claimed to collect, process, and curate more than 17 billion signals from a billion mobile devices daily. The complaint also alleges Gravy Analytics used geofencing to create a virtual geographical boundary to identify and sell lists of consumers who attended certain events related to medical conditions and places of worship. The unauthorized data brokering put consumers at risk of stigma, discrimination, violence, and other harms, according to the complaint. “You may not know a lot about Gravy Analytics, but Gravy Analytics may know a lot about you,” reads a joint statement by FTC commissioners Alvaro M. Bedoya, Rebecca Kelly Slaughter, Melissa Holyoak, and Khan. Gravy Analytics merged with Unacast last year. The company’s website says it offers “location intelligence for every business.” Mobilewalla’s website says its products “make your AI smarter with high-quality, privacy compliant consumer data and predictive feature …” InformationWeek has reached out to Gravy Analytics and Mobilewalla for comment and will update with any response. source

Just adding a chatbot on your website or mobile app will not reduce customer service calls. Sometimes as an analyst, it is my job to break someone’s heart, such as the times when I must tell a customer service team that it is not surprising that their well-designed chatbot is not reducing customer service calls. In fact, it’s doubtful that they will ever see the return on the investment they promised their managers. Here’s the wacky part: In fact, that bot may be more valuable to the brand than the cost savings that increased customer service automation might have provided. But the remaining problem is that, in their business case, they committed to the wrong success metric. What do you mean, “the wrong success metric”? Aren’t bots supposed to reduce my costs? While there is some overlap, people who want to chat are often not the same people who call customer service. Think about a prospect on your website. If they can’t find an answer to their question, they might be willing to try a chatbot, but they are not likely to dig around to find your 800-number. That prospect may simply go to your competitor’s website, and you just lost a sale or a deal that you didn’t even know was in play. But if that chatbot deftly engages a prospect who ends up buying, that chatbot shows immense value to the brand. That chatbot will not absorb customer service calls, however. Make the best of a bad situation: Use chat for call deflection. If your goal is to move more customer service interactions to digital, use chat to deflect calls. This can be as simple as offering a chat to anyone on hold with customer service and sending a link to your chatbot if they take you up on the offer. If you do this, make sure that your digital support is as high-quality as what you offer by phone. If you send them to digital and your analytics tell you that they call you back within minutes, you’ll know that you have just executed an epic customer-experience fail. Make sure that your chatbot is comprehensive and well designed, and have live agents in place as backups to ensure that any customer issue that isn’t successfully handled in self-service can be solved with a human agent. What if I’m still creating the business case for our chatbot? If you are reading this before you promised your leadership that your chatbot will reduce customer service calls, huzzah! To build the right business case, you need to do some analysis on your web or mobile application users via tools such as journey mapping, voice of the customer, confidence-building measures, and customer analytics: Where (and why) do they get hung up? What are their “pain points”? What can’t they find on your website or in your mobile app? Where is the information that the customer needs to move forward? What questions aren’t you answering throughout the customer journey in your digital touchpoints? With these insights, you can identify what your potential chatbot will do — e.g., generate more sales, assist with payment, or whatever it may be. It’s possible that your customer support site has problems and that people are calling customer service after failing there, but you need to validate that assumption, not build a business case on it. … And analyze whether a voicebot may be a better option. If your domain is customer service and you want to reduce contact center costs, explore an advanced voicebot that automates more calls. A modern voicebot will also reduce agent call durations by ensuring that all calls will be escalated to the best available agent with all the pertinent customer information required to solve the customer’s problem quickly. Many brands look to deploy chatbots first, thinking that they are simpler to deploy and a better experience for customers, but if your goal is to reduce customer service calls, put your bot on the phone. Every call that the bot handles is a call that did not need an agent. If you do go down the voicebot path, think about omnichannel for the long run, meaning you should find a vendor that can provide digital as well as voice services. This way, when you deploy your chatbot, you will provide consistent customer experiences and leverage your workflow and integration work across all interactions, saving development time. Want to discuss your chatbot business case, success metrics, or modern voicebots? Forrester clients can schedule a guidance session or inquiry with me.  source

The Chinese government is investigating U.S. chipmaker NVIDIA for allegedly violating its anti-monopoly law by acquiring interconnect provider Mellanox. On Monday, the State Administration for Market Regulation made a statement via China Central Television announcing the investigation, but it does not discuss the specifics of NVIDIA’s suspected violations. The authority approved NVIDIA’s $6.9 billion acquisition of Mellanox, an Israeli company, in 2020 with certain conditions. These aimed to prevent the tech giant from restricting competition in the markets of GPU acceleration, private internetworking devices, and high-speed Ethernet adapters. SEE: EU Investigates NVIDIA Deal With Run:ai Mellanox was required to provide information about new products to Chinese rivals within 90 days of making them available to NVIDIA and give them a chance to ensure their own products are compatible, according to Bloomberg. Conditions also included prohibitions on product bundling, discrimination against customers who buy products separately, and unreasonable trading terms. A NVIDIA spokesperson told TechRepublic: “NVIDIA wins on merit, as reflected in our benchmark results and value to customers, and customers can choose whatever solution is best for them. “We work hard to provide the best products we can in every region and honor our commitments everywhere we do business. We are happy to answer any questions regulators may have about our business.” More must-read AI coverage Latest shot fired in the U.S.-China chip war The investigation represents just the latest move in the years-long tussle for dominance in the lucrative semiconductor market between the U.S. and China. NVIDIA is the leading provider of artificial intelligence and gaming chips, announcing record revenues of $30 billion (£24.7 billion) in the second quarter of 2024. The U.S. is keen to maintain its current sovereignty by blocking China from access to NVIDIA’s state-of-the-art hardware, which is crucial for running advanced AI models. In addition to financial motivations, the U.S. has also raised concerns about China developing AI for military purposes. In 2022, the U.S. applied its first set of chip-related export controls on the sale of semiconductors to Beijing and separately banned NVIDIA from selling its most advanced chips to Chinese companies. In response, NVIDIA developed the China-specific A800 and H100 chips that were compliant with the new controls, enabling it to maintain customers in the country. That same year, the U.S. passed the CHIPS Act, which provided needed semiconductor research investments and manufacturing incentives and reinforced America’s economy, national security, and supply. It also launched a blueprint for an AI Bill of Rights to help regulate AI domestically. Intel, TSMC, Texas Instruments, and Samsung — the world’s largest memory chipmaker — have all announced plans to build fabs in the U.S. SEE: Global Chip Shortage: Everything You Need to Know Then, in August 2023, China’s Ministry of Commerce enforced export controls on gallium and germanium-related items “to safeguard national security and interests.” These rare metals are essential in chip production, and China produces 98% and 54% of the world’s supply of gallium and germanium, respectively. According to data from the Financial Times, the cost of the minerals has almost doubled in the year since. In October last year, the U.S. imposed a second set of export restrictions on semiconductors, closing some of the loopholes NVIDIA exploited with A800 and H100. Since then, the chips giant has been preparing to release new iterations that bypass the updated rules. Nevertheless, the restrictions have greatly impacted NVIDIA’s earnings in China. The country accounted for just 16.9% of its revenue in 2023, 9.5% less than in 2021, according to its latest financial results. Just last week, the Biden administration announced its third set of restrictions on semiconductor exports to China, expanding the list of banned technologies. Beijing responded with a statement, declaring it a “typical act of economic coercion and non-market practice.” “The US says one thing and does another, constantly generalizing the concept of national security, abusing export control measures, and implementing unilateral bullying,” the Ministry of Commerce spokesperson said. “China firmly opposes this.” In response China swiftly banned the sale of germanium and gallium to the U.S., closing loopholes from its 2023 export controls, and added a number of U.S. defense tech startups that cannot do business in China. Quests for AI sovereignty surging worldwide It’s not just the U.S. and China that want to reduce their reliance on other countries regarding AI chips. Both Japan and the Netherlands have struck deals with the White House to restrict the sale of chipmaking kits to China. The U.K. blocked most license applications for companies seeking to export semiconductor technology to China in 2023. That same year, the U.K. government announced that it would devote £100 million ($126 million) to fostering AI hardware development and shoring up possible computer chip shortages. Amazon Web Services also announced plans to invest £8 billion in data centres in the country over the next five years. SEE: UK Government Announces £32m for AI Projects After Scrapping Funding for Supercomputers The European Union offered €43 billion ($46 billion) in subsidies to boost its semiconductor sector with its European Chips Act, which was adopted in July 2023. The bloc also has the lofty goal of producing 20% of the world’s semiconductors by 2030. Global antitrust investigations into NVIDIA NVIDIA is having trouble mediating the U.S.-China chip wars. In addition to the Beijing investigation, the U.S. Justice Department is investigating whether the company violated its antitrust laws by punishing customers who also buy from its competitors and making it difficult to switch suppliers, according to Bloomberg. SEE: AI Surge Could Trigger Global Chip Shortage by 2026 Benoît Cœuré, the president of the French competition authority, has also said that NVIDIA may face antitrust charges in the country “one day” at a July press conference, Bloomberg has reported. source

Long practiced in emerging companies, product-led growth (PLG) has been touted as one of the fastest ways for B2B firms to grow. Perhaps even more compelling than rapid growth is the lower cost of sales in PLG motions. Because the methodology is based on simplified products targeting individual users for self-serve purchase, initially, there is no need for business development reps or sales outreach. Traditional sales-driven firms, don’t despair! You can still get in on the action by adopting PLG strategies that complement your sales efforts at each stage of the customer growth trajectory — and drive faster and more profitable growth. Here are five steps to drive customer growth with PLG: Step 1: Drive user growth to seed the market. PLG can generate rapid user growth by relying on users to do the selling. Offer free trials and make sure that the product has a network effect where users gain more value the more others are using it — by collaborating within their team (e.g., Jira), across their company (e.g., Slack) or even across companies (e.g., Calendly). Create user referral programs where users are incentivized to share the product. These network and viral effects can drive “exponential growth” across markets and accounts. Step 2: Turn heavy user companies into product-qualified accounts. With users seeded across multiple companies, segments, and even regions, it’s easy for PLG organizations to identify the accounts where more users have adopted their product. Accounts with enough active users become new opportunities in the pipeline for a sales rep to close. This process is typically less costly than traditional top-of-funnel marketing efforts, and these product-qualified accounts are considered to be “better than the best” of traditional pipeline opportunities. Step 3: Leverage product telemetry to optimize the experience and build loyalty. A product that delivers fast time to value is foundational to PLG success and will help drive growth and retention for all selling motions. Build in product analytics so you can pinpoint user friction and optimize the time and effort it takes users to achieve their desired outcomes. This type of product telemetry can be used across small and simple or large and complex software modules and is instrumental in improving the user experience and building ongoing loyalty. Step 4: Use in-product, personalized messaging to upsell customers to higher tiers. In PLG motions, the product is the primary marketing and selling method. Create contextual, personalized messages that both provide tips for specific activities and showcase additional offerings that could extend the value that users receive. In the context of existing workflows, alert users to new features, product extensions, or higher-tiered offerings. Offer trials for premium capabilities to make it easy for users to experience the value before expanding their purchase. Step 5: Combine product- and sales-led efforts to expand into new buying centers. Now that you’ve set up a PLG motion, use it to extend to new buying centers with the support of traditional sellers. Account teams should scout out new buyers and identify new use cases for offerings within accounts. Gain cross-sell business through PLG motions using trials and referral programs to incentivize users to share across buying centers. PLG strategies, while practiced successfully at smaller firms, have become additional arrows in the toolkit of go-to-market practices for many larger B2B firms. Pursuing a bottom-up PLG strategy in conjunction with traditional sales efforts has been shown to have the best results for rapid and scalable revenue growth. Just look at the success of Atlassian, Airtable, Dropbox, Calendly, HubSpot, and others to see how well the PLG and sales combination works. Interested in finding out more about PLG? Read this blog on adopting PLG strategies. Clients can access the reports B2B Companies Must Implement Product-Led Growth Practices To Remain Competitive and Leverage Product-Led Growth Strategies For Customer Acquisition, Retention, And Expansion on the Forrester portal or set up a conversation with me. You can also follow or connect with me on LinkedIn. source

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More My, how quickly the tables turn in the tech world. Just two years ago, AI was lauded as the “next transformational technology to rule them all.” Now, instead of reaching Skynet levels and taking over the world, AI is, ironically, degrading.  Once the harbinger of a new era of intelligence, AI is now tripping over its own code, struggling to live up to the brilliance it promised. But why exactly? The simple fact is that we’re starving AI of the one thing that makes it truly smart: human-generated data. To feed these data-hungry models, researchers and organizations have increasingly turned to synthetic data. While this practice has long been a staple in AI development, we’re now crossing into dangerous territory by over-relying on it, causing a gradual degradation of AI models. And this isn’t just a minor concern about ChatGPT producing sub-par results — the consequences are far more dangerous. When AI models are trained on outputs generated by previous iterations, they tend to propagate errors and introduce noise, leading to a decline in output quality. This recursive process turns the familiar cycle of “garbage in, garbage out” into a self-perpetuating problem, significantly reducing the effectiveness of the system. As AI drifts further from human-like understanding and accuracy, it not only undermines performance but also raises critical concerns about the long-term viability of relying on self-generated data for continued AI development. But this isn’t just a degradation of technology; it’s a degradation of reality, identity, and data authenticity — posing serious risks to humanity and society. The ripple effects could be profound, leading to a rise in critical errors. As these models lose accuracy and reliability, the consequences could be dire — think medical misdiagnosis, financial losses and even life-threatening accidents. Another major implication is that AI development could completely stall, leaving AI systems unable to ingest new data and essentially becoming “stuck in time.” This stagnation would not only hinder progress but also trap AI in a cycle of diminishing returns, with potentially catastrophic effects on technology and society. But, practically speaking, what can enterprises do to ensure the safety of their customers and users? Before we answer that question, we need to understand how this all works. When a model collapses, reliability goes out the window The more AI-generated content spreads online, the faster it will infiltrate datasets and, subsequently, the models themselves. And it’s happening at an accelerated rate, making it increasingly difficult for developers to filter out anything that is not pure, human-created training data. The fact is, using synthetic content in training can trigger a detrimental phenomenon known as “model collapse” or “model autophagy disorder (MAD).” Model collapse is the degenerative process in which AI systems progressively lose their grasp on the true underlying data distribution they’re meant to model. This often occurs when AI is trained recursively on content it generated, leading to a number of issues: Loss of nuance: Models begin to forget outlier data or less-represented information, crucial for a comprehensive understanding of any dataset. Reduced diversity: There is a noticeable decrease in the diversity and quality of the outputs produced by the models. Amplification of biases: Existing biases, particularly against marginalized groups, may be exacerbated as the model overlooks the nuanced data that could mitigate these biases. Generation of nonsensical outputs: Over time, models may start producing outputs that are completely unrelated or nonsensical. A case in point: A study published in Nature highlighted the rapid degeneration of language models trained recursively on AI-generated text. By the ninth iteration, these models were found to be producing entirely irrelevant and nonsensical content, demonstrating the rapid decline in data quality and model utility. Safeguarding AI’s future: Steps enterprises can take today Enterprise organizations are in a unique position to shape the future of AI responsibly, and there are clear, actionable steps they can take to keep AI systems accurate and trustworthy: Invest in data provenance tools: Tools that trace where each piece of data comes from and how it changes over time give companies confidence in their AI inputs. With clear visibility into data origins, organizations can avoid feeding models unreliable or biased information. Deploy AI-powered filters to detect synthetic content: Advanced filters can catch AI-generated or low-quality content before it slips into training datasets. These filters help ensure that models are learning from authentic, human-created information rather than synthetic data that lacks real-world complexity. Partner with trusted data providers: Strong relationships with vetted data providers give organizations a steady supply of authentic, high-quality data. This means AI models get real, nuanced information that reflects actual scenarios, which boosts both performance and relevance. Promote digital literacy and awareness: By educating teams and customers on the importance of data authenticity, organizations can help people recognize AI-generated content and understand the risks of synthetic data. Building awareness around responsible data use fosters a culture that values accuracy and integrity in AI development. The future of AI depends on responsible action. Enterprises have a real opportunity to keep AI grounded in accuracy and integrity. By choosing real, human-sourced data over shortcuts, prioritizing tools that catch and filter out low-quality content, and encouraging awareness around digital authenticity, organizations can set AI on a safer, smarter path. Let’s focus on building a future where AI is both powerful and genuinely beneficial to society. Rick Song is the CEO and co-founder of Persona. DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation. If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers. You might even consider contributing an article of your own! Read More From DataDecisionMakers source

By Nadia Dreid ( December 11, 2024, 9:15 PM EST) — For all their disagreements, Republicans and Democrats were largely singing the same tune Wednesday afternoon at a Senate hearing on the security of the nation’s communications networks — that they’re worried, and the government needs to get to work on a solution…. Law360 is on it, so you are, too. A Law360 subscription puts you at the center of fast-moving legal issues, trends and developments so you can act with speed and confidence. Over 200 articles are published daily across more than 60 topics, industries, practice areas and jurisdictions. A Law360 subscription includes features such as Daily newsletters Expert analysis Mobile app Advanced search Judge information Real-time alerts 450K+ searchable archived articles And more! Experience Law360 today with a free 7-day trial. source

Despite cynical theories of slowdown, the Big 4 software giants’ growth is being rocket-fuelled by cloud migrations, as their recent quarterly reports show. But could the looming “service software” AI tsunami threaten the very business they are trying to sustain? source