Latest news with #LLaMA
Yahoo
4 days ago
- Yahoo
Lawyers Just Discovered Something About Meta's AI That Could Cost Zuckerberg Untold Billions of Dollars
A legal expert found that Meta's AI is able to spit out entire portions of books verbatim — and if he's right, it could be seriously bad news for the company and its CEO Mark Zuckerberg. First, a quick primer. All the AI that's commercially buzzy at the moment, like OpenAI's ChatGPT or Meta's Llama, is trained by feeding in huge amounts of data. Then researchers do a bunch of number crunching using algorithms, basically teaching the system to recognize patterns in all that data so thoroughly that it can then create new patterns — meaning that, say, if you ask for a summary of the plot of one of the "Harry Potter" books, it'll give you (hopefully) a reasonable overview. The problem, Stanford tech law expert Mark Lemley explains in an interview with New Scientist, is that his team's research found that Meta's LLaMA is able to repeat verbatim the exact contents of copyrighted books — such as, in one example he found, lengthy passages from the multi-billion dollar "Harry Potter" series. For Meta, this is a gigantic legal liability. Why? Because if its AI is producing entire excerpts of material used to train it, it starts to look less like its AI is producing transformative works based on general patterns about language and the world it learned from its training data, and more like the AI is acting like a giant .ZIP file of copyrighted work, which users can then reproduce at will. And it looks a lot like it is. When testing out various AI models by companies including OpenAI, DeepSeek, and Microsoft, Lemley's team found that Meta's LLaMA was the only one that spat out book content exactly. Specifically, the researchers found that LLaMA seemed to have memorized material including the first book in J.K. Rowling's "Harry Potter" series, F. Scott Fitzgerald's "The Great Gatsby," and George Orwell's "1984." It's not under debate that Meta, like its peers in the tech industry, used copyrighted materials to train its AI. But its specific methodology for doing so has come under fire: it emerged in copyright lawsuit against Meta by authors including the comedian Sarah Silverman that the model was trained on the "Books3" dataset, which contains almost 200,000 copyrighted publications and which Meta engineers downloaded using an illegal torrent ("Torrenting from a [Meta-owned] corporate laptop doesn't feel right," one of them fussed while doing so, in messages produced in court.) Lemley and his team estimate that if just three percent of the Books3 dataset were found to be infringing, the company behind it could owe nearly $1 billion in statutory damages, and that's not counting any additional payouts based on profits gleaned from such theft. And if the proportion of infringing content is higher, at least in theory Meta could end up nailed to the wall. Lemley is in a weird position, by the way. He previously defended Meta in that same lawsuit we mentioned above, but earlier this year, the Stanford professor announced in a LinkedIn post that he would no longer be representing the company in a protest of Meta and Zuckerberg's right-wing virtue signaling. Back then, he said he believed Meta should win its case — but based on his new research, it sounds like that opinion may have shifted. Meta declined to comment to New Scientist about Lemley's findings. More on Meta: Meta Says It's Okay to Feed Copyrighted Books Into Its AI Model Because They Have No "Economic Value"
International Business Times
14-06-2025
- Business
- International Business Times
Who is Alexender Wang? 28-year-Old, 'Scale AI' CEO Chosen to Lead Meta's $14.3B 'Superintelligence' Bet
In a major move, technology giant Meta has not only acquired a 49% stake in Scale AI by investing $14.3 billion but has also recruited its 28-year-old CEO, Alexandr Wang, to lead Meta's superintelligence unit. This marks a shift in priorities for artificial intelligence development. This is not a regular AI top talent hiring by Meta, as Wang, who dropped out from MIT to build his own AI empire, is not known for his academic excellence but has a reputation for operational execution in his role as one of the two cofounders of Scale AI. His company made its name by mobilizing large networks of human data annotators—through platforms like Remotasks—to train machine learning systems. With this acquisition, Meta is signaling that owning the data "pipes," rather than just the model architectures, is the real power play in the AI arms race. While Meta's competitors Google and OpenAI are focusing on refining the algorithm, Mark Zuckerberg's firm is now strategically focusing more on owning the entire AI lifecycle—from data generation to model training and product deployment. This vertical integration has parallels to the way companies such as Apple control hardware and software to create tighter feedback loops and promote faster innovation. Meta, once a pioneer in open-source models, such as LLaMA, has faced delays in its AI roadmap and talent drain in its key teams in recent times. Bringing in Wang is interpreted as further indication that the company is moving towards a more product-oriented approach to superintelligence, like Sam Altman opted for with OpenAI. The company is betting that this approach of strategic leadership and scalable data operations will outpace the academic-style development of models. The investment values Scale at $29 billion and comes just weeks after a previous funding round—backed by Nvidia and Amazon—that had valued the company at $14 billion. It also marks Meta's second-largest acquisition, following its $19 billion purchase of WhatsApp. With Wang's recruitment immediately after investing in Scale AI, Meta intends to show its serious intent in the supremacy race of AI, with players like Google DeepMind, OpenAI, and China's DeepSeek leading the charge.
Int'l Business Times
11-06-2025
- Business
- Int'l Business Times
Trump Admin's Plans to Push AI Across Government Sites Leaked on Code Sharing Website
The Trump administration's plan to integrate artificial intelligence across federal agencies has been exposed through a leaked draft of a government-run website, revealing an initiative set to launch on July 4 that would track and promote AI use across departments. The early details were uncovered in code uploaded to GitHub by the General Services Administration's Technology Transformation Services (TTS), led by former Tesla engineer Thomas Shedd, according to 404 Media. The website, is described as a centralized platform offering integration with AI tools from OpenAI, Google, Anthropic, AWS Bedrock, and Meta's LLaMA. It also includes an analytics feature that will reportedly measure AI adoption rates by specific government teams. The project is part of a broader push by Shedd and the Department of Government Efficiency, spearheaded by Elon Musk, to rapidly embed AI technologies into government operations. Leaked audio from a TTS meeting in February revealed that Shedd wanted AI tools to write software, review contracts, and standardize usage across agencies—goals that internal staff reportedly viewed with widespread skepticism. Concerns raised by government employees include the potential for AI-generated code to introduce security flaws, create software bugs, or mistakenly recommend cancelling essential contracts. Despite these warnings, the GitHub page suggests that the initiative is moving forward, with set to launch on Independence Day. As of now, redirects to the White House homepage, and the staging version of the site is hosted quietly on The GSA has not commented publicly on the leak or the concerns surrounding the project. Originally published on Latin Times
India Today
11-06-2025
- Business
- India Today
The US government is building its own AI chatbot with help from a former Tesla engineer
The United States government is preparing to launch its own AI chatbot and integration platform on July 4 under the name according to a report by 404 Media, which found a related code posted on GitHub. The initiative, which aims to 'accelerate government innovation with AI,' is being developed by the General Services Administration's Technology Transformation Services (TTS), headed by Thomas Shedd, a former Tesla engineer. The project includes a website, a chatbot, and an application programming interface (API) that will allow government agencies to tap into AI models developed by OpenAI, Google, Anthropic, and eventually Amazon Web Services' Bedrock and Meta's LLaMA, according to code and early website drafts uncovered by 404 early version of the homepage, which currently redirects to reportedly advertises: 'Three powerful AI tools. One integrated platform.' These include the AI assistant chatbot, a model-agnostic API, and a console to monitor how government teams are using AI. The system will also feature analytics showing usage levels across various Shedd has been at the forefront of the US government's growing interest in artificial intelligence. According to leaked internal meetings and previous public remarks reported by 404 Media and Wired, Shedd wants to 'AI-ify' large parts of federal operations. 'We want to start implementing more AI at the agency level and be an example for how other agencies can start leveraging AI,' Shedd reportedly told his team. He added that tools like AI coding agents – which would write software for federal use – and contract analysis systems are among the first products in development. The broader goal, according to Shedd, is to build centralised AI solutions that federal agencies will eventually be expected to platform appears to be a continuation of ideas proposed under the now-defunct Department of Government Efficiency (DOGE), a short-lived government initiative that was led by Elon Musk recently distanced himself from the Trump administration following a falling out, the influence of DOGE is still evident in current federal tech projects. During its existence, DOGE aimed to reduce bureaucracy and costs by replacing some federal roles with AI-driven it will workWhile the exact functionality of the AI chatbot has not yet been detailed, the underlying API will allow agencies to access a range of AI models and services through a single platform. According to GitHub documentation, integration is being tested on and the platform is still in a staging environment as of early analytics console, also part of the package, will reportedly give visibility into AI usage at each agency. This could potentially help identify which teams are adopting AI effectively, and which may require additional support or training. According to the report, the early version of the platform does not appear to use generic placeholder text, suggesting development is well underway and being tailored for specific government use cases. Tune In
Time Business News
04-06-2025
- Business
- Time Business News
7 Powerful Ways Brands Are Using Generative AI for Ad Copy and Creative
In an era where attention is the most coveted currency, brands are racing to craft ad content that not only grabs eyeballs but drives action. Enter generative AI—an advanced technological marvel reshaping how creative assets and advertising messages are developed. From performance-driven copy to dynamic visuals, brands are now turning to AI to scale faster, personalize deeply, and reduce the manual grind traditionally required to test and optimize ad content. This article explores how using generative AI is no longer just an experimental buzzword in the creative domain but a mainstream strategy that's transforming digital advertising workflows. Whether you're a marketer at a startup or a strategist in a global agency, understanding this shift is no longer optional—it's imperative. Generative AI refers to models that can create new content—text, images, videos, or even audio—based on patterns they've learned from massive datasets. OpenAI's GPT models, Google's Gemini, Meta's LLaMA, and tools like Midjourney and Adobe Firefly have democratized creativity, enabling marketers to conceptualize and launch campaigns at lightning speed. The creative process, once ruled by subjective intuition, now benefits from data-backed assistance that aligns closely with consumer behavior and preference. Major brands are already leveraging AI-driven platforms to develop thousands of ad variations in minutes, test copy performance across segments, and fine-tune messaging for every audience touchpoint. AI isn't replacing creativity—it's enhancing it, making it more scalable and measurable. What makes this technology a game-changer is its ability to analyze, generate, and optimize simultaneously. Traditional A/B testing relies on limited versions and cycles that take time. Generative AI can produce dozens of ad variants and predict which will perform best before you even launch them. For example, eCommerce brands often need seasonal campaign copy tailored to dozens of product categories. Instead of briefing multiple writers or agencies, marketers can now feed key product features, brand tone, and campaign goals into AI tools and get polished drafts ready for market in hours. These systems can adjust tone—whether witty, professional, urgent, or empathetic—based on the campaign's objective and intended audience. AI also enables rapid multilingual adaptation, crucial for global campaigns. Localization no longer requires huge translation teams, as models can translate and culturally adapt messaging while maintaining emotional impact. Generative AI is revolutionizing more than just words. Tools like DALL·E, Runway, and Adobe's suite now allow marketers to create ad visuals from scratch or enhance existing ones. Need a lifestyle image featuring your product on a tropical beach? With generative AI, you can design that image in seconds—no expensive photoshoots required. Fashion retailers have begun using AI to generate lookbooks based on trends and customer preferences. Food and beverage companies are creating photorealistic images of new product concepts for test campaigns. And instead of generic stock images, brands are able to build unique visual assets that resonate more authentically with their identity. Another breakthrough is the creation of short video ads. Platforms like Synthesia or Pika allow marketers to turn scripts into talking-head videos using AI avatars. This unlocks content production at a pace and cost unimaginable just a few years ago. Marketers have long talked about the 'segment of one,' but executing that vision was nearly impossible—until now. Generative AI enables real-time personalization where different ad versions are shown based on user behavior, interests, and past interactions. Dynamic creative optimization (DCO), fueled by AI, can now pair personalized copy and visuals at scale. If one user previously searched for eco-friendly products, they might see a headline emphasizing sustainability. Another user interested in deals will see price-based benefits. AI tools like Persado and are making this real for email subject lines, display ads, social captions, and more. The result is an enormous uplift in engagement rates. One fintech startup reported a 34% increase in click-through rates after switching to AI-personalized ad copy over their standard messaging templates. Traditionally, A/B testing required time and traffic to declare a winner. With AI, marketers can instantly simulate how different ad copies might perform, based on similar campaigns and historical data. AI can predict performance before spending a single dollar on media. Marketers using tools like Jasper or ChatGPT for ideation and testing are reporting higher creative throughput. They can test different emotional tones—curiosity, urgency, trust—across ads to identify which emotional triggers drive the best outcomes. Once live, AI analytics platforms interpret real-time results and suggest actionable optimizations, cutting campaign costs and boosting ROI. Moreover, AI can flag fatigue in ad creative. If click-through rates dip, the system might recommend variations, new formats, or different CTA (call-to-action) placements—automatically improving the campaign lifecycle. While the possibilities of using generative AI in advertising are exciting, they also raise questions. Is AI-generated content truly original? How do we ensure diversity, avoid stereotyping, and maintain authenticity when machines drive creativity? Brands must treat generative tools as collaborators, not creators. Human oversight remains crucial to ensure brand voice, inclusivity, and legal compliance. AI models may not always understand the nuance of sensitive topics or evolving cultural contexts, so marketers should apply judgment and empathy during the final review. Transparency is another factor. Consumers are beginning to question if ads are created by humans or machines. Brands that embrace transparency and disclose their use of AI may even earn greater trust in a world where authenticity is increasingly valued. As AI becomes more integrated into the creative pipeline, marketing teams will need to upskill and adapt. The most successful professionals will be those who combine human creativity with AI's analytical power. Prompt engineering—the art of giving AI the right instructions—will become a core skill, alongside strategic oversight and creative editing. Organizations investing in the Best AI Marketing Course options are giving their teams a competitive edge. These courses cover not just the tools but also the frameworks and ethical standards for responsible AI use in marketing. Learning how to co-create with AI will be the defining skill of the next-gen marketer. Creative directors will shift from being originators to curators and conductors—shaping the vision, while letting AI handle executional tasks. Copywriters and designers will spend more time on strategy, storytelling, and final polish rather than first drafts and mockups. One global apparel brand adopted a generative AI suite to support its launch of a new streetwear collection. Their challenge was to create hyper-local campaigns across 12 countries, each with different cultural aesthetics and product popularity. Using an AI platform trained on past campaign performance and cultural cues, they generated region-specific slogans, product descriptions, and Instagram captions in under 48 hours. Visual assets were created using an AI image generation tool to feature different models, environments, and styles that reflected each market. The result? A 41% increase in engagement compared to their previous campaign, a 25% faster campaign rollout, and a 19% cost reduction in creative production. This success is just one of many across industries ranging from beauty to finance, where generative AI is helping brands meet the demand for fresh, tailored, high-converting ad content. We are witnessing a pivotal shift. What started as experimentation with AI-generated text and images has now become a foundational component of modern advertising. Using generative AI allows for unprecedented speed, personalization, and experimentation—enabling brands to connect more deeply and efficiently with their audiences. However, the winners in this space won't be those who rely blindly on automation. The true leaders will be the marketers and creatives who understand how to guide AI tools with clarity, empathy, and ethical intention—enhancing, not replacing, the human touch. For marketers ready to take the leap, mastering generative AI is not a luxury—it's a necessity. Whether through formal training or hands-on experimentation, now is the time to evolve your skill set and embrace a future where creativity is not just powered by data, but elevated by it. TIME BUSINESS NEWS
