live AI search engine, new AI search tool, AI search engine vs Google, search engine 2025, alternative to Google search, next-gen search technology, AI-powered search engine 2025, smarter AI search, real-time AI search, AI search engine features, Google competition search, AI search results, AI search engine explained, search with AI, AI tools for search, new search engine breakthrough, AI internet search, future of search engines, AI and Google, AI search technology review
Meta is developing its own AI-powered search engine to reduce dependence on Google and Bing, aiming for real-time responses across platforms like Instagram and WhatsApp. The Open Source Initiative (OSI) recently updated its definition of “open” artificial intelligence, challenging Meta’s Llama model, which lacks full transparency on training data and usage restrictions. As OSI pushes for greater transparency, Meta and other tech giants face rising pressure to align with open-source standards, highlighting a pivotal shift in the AI and tech landscape.
🔍 Key Topics Covered:
Meta's AI Search Engine: Meta’s strategic move to compete with Google and Bing
How Meta’s AI integration in apps like Instagram and WhatsApp will offer real-time information
OSI’s new standards for "open" AI, challenging Meta's Llama model over transparency and data access
🎥 What You’ll Learn:
How Meta’s AI search could redefine information flow across its platforms and empower Meta AI
The growing clash between open-source advocates and tech giants over AI transparency
Potential impacts on user experience, with Meta aiming for control over real-time, in-house data
📊 Why This Matters:
This video delves into Meta’s ambitious AI search project and the evolving landscape of open-source AI, exploring the power shifts in tech as companies race for control over information. With Meta, Google, and Microsoft in the spotlight, AI development is transforming the way we search, interact, and consume information online.
DISCLAIMER:
This video explores Meta’s new AI search engine and the broader implications of OSI’s updated open-source standards. It’s essential viewing for anyone interested in AI, tech innovation, and the dynamic between transparency and control in today’s digital landscape.
#ai
#meta
#AISearchEngine #NewSearchEngine #GoogleAlternative #AIinSearch #NextGenSearch #AIpoweredSearch #GoogleVsAI #SearchEngineRevolution #AI2025 #FutureOfSearch #TechNews #AItools #SmartSearch #AItechnology #OpenAIsearch #SearchRevolution #AIinTech #FutureOfInternet #NewTechBreakthrough #AIforBusiness #AIsearch2025
Meta is developing its own AI-powered search engine to reduce dependence on Google and Bing, aiming for real-time responses across platforms like Instagram and WhatsApp. The Open Source Initiative (OSI) recently updated its definition of “open” artificial intelligence, challenging Meta’s Llama model, which lacks full transparency on training data and usage restrictions. As OSI pushes for greater transparency, Meta and other tech giants face rising pressure to align with open-source standards, highlighting a pivotal shift in the AI and tech landscape.
🔍 Key Topics Covered:
Meta's AI Search Engine: Meta’s strategic move to compete with Google and Bing
How Meta’s AI integration in apps like Instagram and WhatsApp will offer real-time information
OSI’s new standards for "open" AI, challenging Meta's Llama model over transparency and data access
🎥 What You’ll Learn:
How Meta’s AI search could redefine information flow across its platforms and empower Meta AI
The growing clash between open-source advocates and tech giants over AI transparency
Potential impacts on user experience, with Meta aiming for control over real-time, in-house data
📊 Why This Matters:
This video delves into Meta’s ambitious AI search project and the evolving landscape of open-source AI, exploring the power shifts in tech as companies race for control over information. With Meta, Google, and Microsoft in the spotlight, AI development is transforming the way we search, interact, and consume information online.
DISCLAIMER:
This video explores Meta’s new AI search engine and the broader implications of OSI’s updated open-source standards. It’s essential viewing for anyone interested in AI, tech innovation, and the dynamic between transparency and control in today’s digital landscape.
#ai
#meta
#AISearchEngine #NewSearchEngine #GoogleAlternative #AIinSearch #NextGenSearch #AIpoweredSearch #GoogleVsAI #SearchEngineRevolution #AI2025 #FutureOfSearch #TechNews #AItools #SmartSearch #AItechnology #OpenAIsearch #SearchRevolution #AIinTech #FutureOfInternet #NewTechBreakthrough #AIforBusiness #AIsearch2025
Category
🤖
TechTranscript
00:01Meta is taking a strategic leap in AI by developing its own search engine with the goal of moving away from relying on Google and Bing for real-time information.
00:11This effort isn't just about wanting control, it's about redefining how information flows within Meta's ecosystem, like on Facebook, Instagram, and WhatsApp.
00:20Meta's current setup leans on two major players for search, Google, with its AI model Gemini, and Microsoft's Bing.
00:27Google's been pushing to make its search more interactive, delivering conversational results through Gemini.
00:32Meanwhile, Microsoft powers ChatGPT's web browsing with Bing, allowing it to pull in live information.
00:39Both of these giants are Meta's competitors, which complicates relying on them for search.
00:43The plan is to integrate Meta's AI search directly with its chatbot, Meta AI, already embedded in popular platforms like Instagram and WhatsApp.
00:52This new search engine will have real-time responses, giving Meta the ability to handle queries about current events, financial markets, and sports.
01:00This move not only ensures Meta's independence, but also strengthens their AI's knowledge base.
01:05Months of groundwork have gone into this, with Meta's web crawlers actively gathering data from around the Internet.
01:11Meta's been collecting everything from general information to location-based data, the kind that could even compete with Google Maps.
01:18For about eight months, this data has been quietly building the foundation of Meta's new AI search engine.
01:23And recently, Meta took another big step by forming a multi-year partnership with Reuters.
01:27This deal means Meta's AI can access writer's content, ensuring it delivers real-time, accurate answers about current events, especially for US-based users across its platforms.
01:37This shift is notable, especially since Meta previously reduced its focus on news due to controversies going so far as to remove the news tab.
01:46Now, through AI, they're doubling down on providing reliable, up-to-date news.
01:50In the bigger picture, Meta's move shows a clear strategy for building its own search framework, potentially developing a system that can stand on par with Google or Bing.
01:58Creating a web crawler that can rival these giants is no small task.
02:03It involves huge infrastructure and advanced machine learning models.
02:07But the payoff is clear.
02:09Control over its own AI-powered answers and possibly even exclusive information Meta might not have if it continued relying on outside partnerships.
02:18There's an added dimension here, too.
02:20Meta's AI will evolve as it gathers data from around the web, gaining insights directly through its own crawlers.
02:26This provides an advantage because the AI can deliver answers not only tailored to Meta's platforms, but also shaped by data Meta itself owns.
02:36And in practical terms, it safeguards Meta if Google or Microsoft, competitors on the AI front, ever decide to end their partnerships.
02:45It's worth noting that Meta isn't alone in this AI search race.
02:49OpenAI is rolling out its own search GPT, while companies like Perplexity are launching similar AI-driven search engines.
02:56But they're facing legal challenges.
02:58Publishers like News Corp and The New York Times are suing over copyright issues, arguing these AI tools pull data without fair compensation.
03:06Meta's setup, especially with its Reuters deal, avoids some of this conflict by ensuring direct access to high-quality, licensed content.
03:14Meta's focus on in-house data gathering and reducing third-party reliance may set them apart in this crowded space.
03:21The direction is clear.
03:22A robust search tool that powers Meta's real-time AI responses across apps, and possibly a broader tool for users that leverages their massive dataset.
03:32There's also a chance this could enhance Meta's other platforms, such as Threads, where real-time insights could bring users up-to-the-minute information on evolving news stories.
03:41Alright, with Meta ramping up efforts to control its own data flow, it highlights how important it is to keep your personal info secure in today's digital world.
03:49And on that note, it's a good time to address the ever-growing online threats that go hand-in-hand with advancements in AI.
03:56See, every time you connect to the internet, whether you're streaming, shopping, or just browsing, your personal info is at risk.
04:04Cybercriminals are constantly looking for ways to steal your data through phishing scams, data breaches, and even man-in-the-middle attacks where they intercept your private communications.
04:13But there's an easy way to protect yourself – NordVPN.
04:17NordVPN uses military-grade encryption to lock down your data, keeping it safe from hackers, snoopers, and malicious websites.
04:24They've got this amazing feature called Threat Protection Pro that block ads and trackers.
04:29But even more importantly, it actively scans your downloads for malware and blocks harmful websites before you even click on them.
04:36You can also monitor the dark web for leaks of your personal info with their dark web monitor.
04:40And get this, NordVPN can protect up to 10 devices at the same time, which means your laptop, phone, tablet, and more can all stay secure wherever you go.
04:49Now here's the best part.
04:50Right now, NordVPN is running an exclusive offer just for AI Revolution viewers.
04:55Get exclusive NordVPN deal plus 4 months extra on this link.
05:00NordVPN.com slash AI Revolution.
05:03It's risk-free with Nord's 30-day money-back guarantee.
05:07Seriously, don't wait until something happens.
05:09Protect yourself today and get NordVPN through the link in the description.
05:14Now, as Meta works to control its own search engine, there's a bigger conversation happening around transparency in AI.
05:20The Open Source Initiative, or OSI, just set new standards on what it really means to be open in AI, putting pressure on tech giants, including Meta, to rethink how open their models truly are.
05:31So, the OSI just put out a big update that's causing waves in the tech world.
05:35They've released a new official definition for what it means to have open artificial intelligence, and it's directly challenging some of the biggest players in the field.
05:43Meta, for instance, has promoted its AI model LAMA as open source, but under OSI's new standards, LAMA doesn't quite make the cut.
05:51For years, OSI has set the standard on what open source should look like.
05:55Their definitions have shaped the industry, guiding developers to work openly, freely sharing their work to encourage collaboration.
06:03Now, with AI coming into play, OSI is expanding its standards.
06:08They're saying that to be truly open, an AI model has to reveal the data used for training, allowing others to understand and potentially replicate it.
06:17This also means making the entire codebase accessible so users know exactly how the AI was built and operates.
06:24On top of that, OSI insists that the specific settings and weights, which fine-tune the AI's responses, need to be available.
06:32These details give the AI its unique performance characteristics, and full access would mean anyone could recreate, modify, or build upon it with the same quality.
06:42For Meta's LAMA, this poses a problem. While LAMA is available for anyone to download, it doesn't allow unlimited commercial use, especially for applications with massive audiences, and it doesn't disclose the training data used.
06:56OSI argues that for something to be open source, it has to be completely open, without restrictions on use or sharing.
07:03Meta's stance is that there's no single, universally agreed definition of open source AI, pointing out the challenges in applying traditional open source rules to today's complex AI models.
07:14Faith Ayshin, Meta's spokesperson, has noted that Meta does agree with OSI on many points, but this new definition goes too far in a way that doesn't account for the nuances of AI development today.
07:26Meta's position is that while they're committed to making AI more open and accessible, they're also mindful of the risks involved in exposing certain details, like training data.
07:37In a sense, this clash echoes the open source battles of the 1990s.
07:41OSI's executive director, Stefano Maffuli, even compared Meta's resistance to Microsoft's initial stance against open source software decades ago.
07:50Back then, Microsoft argued that open source software was a threat to their business model, a similar stance to what Maffuli now sees with Meta.
07:58When he met with Meta's representatives, they talked about the resources poured into Lama's development and questioned if anyone else could replicate that kind of effort.
08:06Maffuli views it as a familiar strategy, using costs and technical complexity as reasons to keep their innovations tightly guarded.
08:13OSI's new definition didn't happen overnight, it was a two-year process involving global experts across various fields, academics specializing in machine learning, philosophers, open source advocates, and even creatives from the creative commons world.
08:26They aimed to create a robust framework that could withstand challenges from major corporations while staying true to the core values of openness and accessibility.
08:35Other leaders in the open source world are on board with OSI's stance.
08:40Clement DeLange, CEO of Hugging Face, has called this new definition a big step forward, especially in addressing the role of training data, which is often kept hidden due to its value as intellectual property.
08:52DeLange believes this transparency is crucial, and he sees OSI's standards as essential for shaping a more open, ethical AI landscape.
09:00Meta's reasons for holding back its training data go beyond just business interests.
09:04They cite safety concerns as well.
09:06However, critics see another angle, arguing that keeping training data private allows Meta to minimize its legal risks and preserve a competitive edge.
09:15Many AI models, Meta's included, are likely trained on copyrighted material, a fact that has sparked numerous lawsuits.
09:21The New York Times reported in April that Meta has internally acknowledged the difficulty of excluding copyrighted content from its training data simply because there's no surefire way to filter it out.
09:33Legal cases against Meta, OpenAI, and others continue to emerge, but it's tough for plaintiffs to prove copyright violations without access to exact data.
09:42A rare exception here is stable diffusion, which openly reveals its training sources, making it easier for copyright holders to understand where their work has been used.
09:52Simon Willison, a researcher and creator of the open source tool, Dataset, sees OSI's new definition as a way to call out what he refers to as open washing, where companies claim their AI is open source without really meeting the standards.
10:06Now that OSI has laid down these ground rules, it could push back against tech companies that, in Willison's view, are using the term open source as a marketing tool rather than a commitment to genuine transparency.
10:18With OSI's updated standards, the tech industry faces a choice, align with these principles or find new ways to justify the secrecy around AI's inner workings.
10:28The Linux Foundation has also joined this debate, recently announcing its own efforts to define open source AI, showing that the conversation is expanding.
10:37It's a critical moment, as the industry decides whether traditional open source values, openness, freedom to modify, and shared development can survive in the rapidly evolving AI landscape.
10:48So, what do you guys think about all this? Let me know in the comments.
10:51And as always, if you enjoyed this breakdown, don't forget to like and subscribe for more on the latest in AI and tech.
10:58Thanks for watching, and I'll see you in the next one.