AI News You Should Know About | Weekly AI Insights

Ashley Coffey & Daniel Hill | AI Technology News

Stay informed with AI News You Should Know About, your weekly briefing on the latest AI news and AI technology trends. Join hosts Ashley Coffey & Daniel Hill as they explore key AI headlines, breakthroughs, and the business impact of AI, helping you stay ahead in the ever-evolving tech landscape in 15 minutes or less.

  1. 1日前

    NVIDIA Buys Hugging Face: What It Means for Open Source AI Tools - AI News

    NVIDIA has agreed to acquire Hugging Face, the online developer platform and open source AI model and dataset library, for nearly $13 billion. The deal, long rumored, became official Thursday. Ashley breaks down what it signals: NVIDIA committing to the software side of generative AI, not just chips, and promising to maintain Hugging Face's open standards. Her take: this is NVIDIA de-risking its business ahead of an eventual cooldown in chip demand, locking in relevance as the platform where developers discover and build models. The open question is whether "open standards" holds up in 18 months, or quietly becomes the on-ramp to NVIDIA's own ecosystem. Then: the Justice Department has filed court papers backing OpenAI's argument that training AI models on internet content is fair use, the first time the federal government has weighed in directly on an AI copyright case, and it's siding against the New York Times in the paper's suit against OpenAI and Microsoft. The DOJ leans on two claims: that the public benefits of training outweigh the harm to publishers, and a national security argument that AI helps analysts process information faster. The Times argues OpenAI used billions of dollars of journalism without permission or payment, and Sarah Silverman and other publishers and authors have joined the suit. Ashley's take: a government brief in a private case isn't neutral, it's the administration picking a side, and the national security argument doesn't answer whether OpenAI paid for what it used. AI news you should know about, in under 10 minutes. In this episode: (00:04) This week's headlines: NVIDIA/Hugging Face and the OpenAI/NYT case (00:37) NVIDIA's $12.9B acquisition of Hugging Face (01:29) What the deal signals about NVIDIA's AI strategy (01:52) Ashley's take: NVIDIA de-risking its business (02:44) Trump administration backs OpenAI in the NYT copyright case (03:08) The DOJ's two-part argument (03:56) The New York Times' response and who else joined the suit (04:23) Ashley's take: fair use doesn't answer whether OpenAI paid for it (04:48) Wrap-up and where to follow the show Links:  Nvidia’s Hugging Face Acquisition Is a $12.9 Billion Bet on Open-Source AI (Wired)Trump administration backs OpenAI in New York Times’ copyright case over training of chatbots (Boston Globe)Follow Ashley at @ashleyrcoffey89 on Instagram and Threads. Send us a topic you want covered, and follow the show for AI news you should know about, in under 10 minutes!  Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

  2. 3日前

    Apple's New 2NM Chip and Why Google Is Cracking Down on Android Apps— AI News

    Apple unveiled its 2-nanometer M6 chip and quad-die M5 Ultra this week — its first chip built on a 2NM process and its most powerful ever. Ashley breaks down what that actually means: a new three-tier CPU layout, dual neural engines, and up to 4.5x more GPU compute, all heavily optimized for running AI locally instead of in the cloud. Daniel raises the question a lot of Mac users are asking — how much of Apple's on-device AI is really just Siri leaning on Gemini under the hood. Then: as the AI data center boom drains the global supply of memory chips, Google is stepping in with new Android app quality requirements, taking effect February 2027, aimed at cutting memory usage and forcing code optimization. Daniel traces the trick developers have used for decades to work around tight memory (yes, the original Xbox is involved), and Ashley makes the case that this isn't just a performance play — with a large base of older, low-memory phones still in use, it's a safety move to stop apps from overloading devices and overheating batteries. AI news you should know about, in under 10 minutes! In this episode: (00:35) Apple's M6 and M5 Ultra chips explained (01:17) What "2NM" actually means for on-device AI (01:47) Local AI vs. the cloud — and how much of it is really Gemini (02:51) AI's memory crunch is coming for Android apps (03:32) The Xbox-era trick developers use to work around limited memory (03:58) Google's new app requirements, effective February 2027 (04:32) Why this is really a safety issue, not just a performance one Links:  Apple unveiled its 2nm M6 and quad-die M5 Ultra chips, promising up to 4.5x more GPU compute for on-device AI. (Apple) AI’s memory crunch is coming for Android apps (TechCrunch) Follow Daniel at @danielhillmedia on Instagram and Threads. Follow Ashley at @ashleyrcoffey89 on Instagram, and grab the free AI newsletter linked in her bio. Want to hear about a specific topic? Send us a DM! Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

  3. 8月28日

    Meta's $18B Settlement and the AI Copyright Question Nobody's Answering

    Meta agreed to an $18 billion multi-state settlement over claims it designed Facebook and Instagram to be addictive and harmful to kids. The financial penalty is the headline, but the real story is buried in the terms: Meta is offering $5.3 billion to the 29 states in the suit (Oklahoma included) to pursue its competitors. Ashley breaks down why Meta's biggest fear was never the teen account restrictions, it was being the only platform that had them. Then: is it legal to train AI models on copyrighted books? A judge ordered Anthropic to pay $1.5 billion to authors, but ruled the training itself was lawful. The problem was where the books came from. Daniel unpacks the fair use question, why copyright law hasn't been meaningfully updated since 1976, and what this precedent means for independent creators whose work was scraped without permission or payment. AI news you should know about, in under 10 minutes. In this episode: (00:38) Meta's $18B settlement and the teen account restrictions (01:28) The $5.3B incentive to go after TikTok and YouTube (02:37) Beauty filter limits and what they actually cover (03:52) Anthropic's $1.5B payout and the piracy distinction (04:48) Why 1976 copyright law can't handle this (06:53) Sora, scraped creator content, and getting paid zero Links: Show notes include "If Meta's going down, it's taking TikTok and YouTube with it" (The Verge) and the full AI copyright piece.  Follow Daniel at @danielhillmedia on Instagram, Threads, and YouTube. Follow Ashley at @ashleyrcoffey89 on Instagram, and grab the free AI newsletter linked in her bio. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

  4. 8月26日

    Latest AI News: Amazon Destroys Rare Books, Meta AI on Mac

    In this episode (00:38) Meta AI launches a dedicated desktop app for Mac Meta's standalone Mac app targets business owners and creators who run on Facebook and Instagram. Meta AI can now tap into its own analytics and advertising tools, and professional accounts can connect Google Workspace to pull in documents and spreadsheets. Meta says users can ask questions about their business and get answers drawn from account engagement and ad performance data, plus publicly available insights on how comparable brands are building their presence. (01:45) Daniel's take: analytics-aware AI is the real feature Daniel draws the parallel to YouTube's Ask Studio, where you can ask an AI to review your last ten videos and tell you what worked. Being able to ask why a suggestion will work is where the value shows up. His one complaint: no Windows version, at least not yet. (02:51) Why do AI tools keep launching on Apple first? Ashley points to how the platforms are built from the ground up and the different exploit and remediation timelines. Daniel offers a complementary theory: fewer Apple users means fewer bad actors targeting them. Their conclusion is that it is probably multiple compounding reasons. (04:08) 404 Media tracks rare books to an Amazon AI training facility Background: booksellers had reported a historic spike in bulk purchases, made anonymously, in seemingly random selections, by buyers who were not price sensitive at all. Printed books are prized training data because the text was never scraped from the web, it is cleanly organized, and anything printed before 2022 is guaranteed free of AI-generated text. Models trained on AI-generated text get worse. (05:24) Inside the Las Vegas warehouse 404 Media planted a tracking device in a rare book and followed it through multiple stops to an Amazon warehouse in Las Vegas. Employees describe receiving enormous shipments, cutting the bindings off to scan faster, and destroying the books. Barcodes and serial numbers get scanned, consistent with a methodical effort to work through the world's ISBN list. Many of the books are rare, out of print, or in foreign languages, meaning few people would notice their hard copies disappearing. (06:37) Ashley's take: history does not look kindly on this Ashley recommends the 404 Media podcast episode on the investigation (linked below). Her position: digitizing is good, destroying is not, and none of this addresses the fact that creators were never paid for the data sets behind today's models. (07:28) Nobody is paying the authors Daniel agrees that digitization has real preservation value and notes that once a book is sold, the seller loses control of it. The unresolved problem is that the original writer gets nothing while their work trains an LLM. (08:14) Until regulation forces a change, this continues Ashley's closing point on where this goes next. Links 404 Media's investigation into Amazon and the rare book shipments: https://archive.ph/70Hun  404 Media podcast episode on the investigation (Ashley's recommendation): https://pod.link/1703615331/episode/NTlmNDdlNGUtOWI1Ni0xMWYxLTlkYzUtMzcwNmUzNGZmZWMx  Meta's post announcing the Meta AI desktop app: https://ai.meta.com/meta-ai/download/  Connect with us Ashley Coffey: @ashleyrcoffey89 on Instagram and Threads Daniel Hill: @danielhillmedia on Instagram, Threads, and YouTube Follow the show and send us topics you want covered. We do this for you! Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

  5. 8月21日

    OpenAI's ChatGPT Ads Expand to 31 European Markets: What It Means for AI Trust

    ChatGPT ads have officially landed! OpenAI's biggest advertising rollout yet is now live across 31 European markets, targeting Free and Go-tier users. Hosts Ashley Coffey and Daniel Hill break down why OpenAI is leaning on explicit user consent (not "legitimate interest") to stay GDPR-compliant, and why this move is really about revenue ahead of a possible IPO. Then: why doesn't anyone trust AI yet? A new TechCrunch report, backed by comments from Anthropic's CEO, a CNBC trust poll of 18–34 year olds, and a Wall Street Journal piece on AI data center backlash — shows the AI industry has a real trust problem, not just a PR problem. Ashley and Daniel dig into the rise of "dumb tech" and slow-tech trends among younger generations, Airbnb CEO Brian Chesky's comments on the AI backlash, and where AI actually earns its keep (hint: not everywhere).   ChatGPT Ads Land in 31 European Markets (00:35–03:01) Source: Digiday (link in original article) OpenAI's largest ad rollout to date goes live across 31 European markets today, shown only to Free and Go-tier ChatGPT users. Unlike some ad platforms that rely on "legitimate interest" to justify targeting, OpenAI is asking for explicit user consent — a GDPR (General Data Protection Regulation) requirement, since personalized ads count as processing personal data. Context: this rollout comes as OpenAI and Anthropic both position themselves ahead of potential future IPOs — and ad revenue is attractive to investors. Ashley & Daniel's take: Both hosts are skeptical ads will be a viable long-term revenue strategy for AI companies. Ashley flags the trust risk — once ads enter the picture, users start questioning whether an AI's answer is the best one or the highest-paying one. Daniel predicts ad volume will keep climbing if profitability doesn't materialize quickly. AI Was Supposed to Win People Over By Now — It Hasn't (03:11–08:31) Source: TechCrunch (link in original article) Anthropic's CEO posted on X this week acknowledging AI's trust crisis, saying the public suspects AI companies are "cooking up some new way to screw them over" — and admitting the industry hasn't delivered on big promises like curing cancer. A CNBC poll found a majority of 18–34 year olds don't trust top AI industry leaders to act responsibly. The Wall Street Journal reports tech companies are facing a PR crisis over AI data center buildouts across the US, now offering job guarantees and clean water investments to ease community pushback. Consumer trend watch: younger generations are turning to "dumb phones," point-and-shoot cameras, tape decks, CDs, iPods — plus offline hobbies like knitting, quilting, puzzles, and mahjong. In-person activities like run clubs are outpacing online dating in popularity. Airbnb CEO Brian Chesky says the AI backlash is real — not because of messaging, but because the industry isn't shipping products regular people can actually use or afford. Ashley & Daniel's take: Skepticism toward new tech is human nature (see: the Model T vs. horses). Ashley argues this is a leadership problem, not just a PR problem, and calls for more community input on data center siting decisions. Both agree AI has real, valuable use cases — like expanding access to mental health support or medical guidance for people who lack insurance or scheduling flexibility — but isn't a wholesale replacement for human alternatives. Ashley also flags Gen Alpha's growing interest in "low-tech" and slow-tech living as a trend worth watching. Links & Where to Find Us Instagram, Threads & YouTube: @DanielHillMedia Instagram: @AshleyRCoffey89 Get the highlights without the twice-a-week commitment: sign up for Ashley's free monthly AI newsletter — link in her Instagram bio. Got a topic you want covered? Send it to us on social media. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

  6. 8月18日

    Opt Out or Get Trained: Twitch's AI Policy + NVIDIA's Risky $500B Plan

    Twitch Will Train Amazon's AI on Your Content — Unless You Opt Out (00:40) Twitch now trains Amazon's generative AI models on streamers' broadcasts, clips, VODs, chat text, and images by default. Streamers who want out have to manually turn it off: Go to Settings → Security and Privacy in your Twitch account Find the toggle labeled "Training for Generative AI" Turn it off to stop Amazon from using your streams, VODs, clips, highlights, chat, and images for future model training Opting out only blocks generative AI training. Twitch can still use AI for captions, recommendations, sponsorship tools, streamer growth features, monetization, and safety tools like AutoMod. Ashley and Daniel discuss why Amazon didn't simply pay streamers for their data, and how this mirrors the original, uncompensated scraping that built today's AI models in the first place. Source: Twitch will train Amazon’s generative AI on your streams, clips, and chat by default unless you manually opt out. (Apple Insider)   NVIDIA's $500 Billion Plan Is Risky but Brilliant — Especially for Aging GPUs (02:43) NVIDIA has lined up commitments of up to $500 billion from major investment firms — including BlackRock, Blackstone, and Goldman Sachs — to fund new AI data centers. To win their confidence, NVIDIA is personally guaranteeing the resale value of the GPUs backing those deals: if the chips used as loan collateral lose value and a lender has to liquidate, NVIDIA will cover up to 25% of the shortfall. Daniel draws a parallel to Lucent Technologies during the dot-com crash, which financed its own customers' equipment purchases and collapsed when the bubble burst. He also flags the "wrong-way risk" this creates for NVIDIA: its financial obligations grow biggest exactly when demand — and its own revenue — is weakest. Ashley and Daniel also cover CEO Jensen Huang's public comments to investors and bond markets aimed at calming fears about NVIDIA's exposure, and debate whether this financing structure is a bold, necessary innovation or a warning sign of a bubble nearing its tipping point. Source: Nvidia’s new $500B plan is risky but brilliant, especially for aging GPUs (TechCrunch)   Connect With Us Daniel Hill: Instagram, Threads & YouTube — @DanielHillMedia Ashley Coffey: Instagram — @AshleyRCoffey89 Want the monthly digest instead of twice-a-week episodes? Ashley's AI newsletter link is in her Instagram bio. Got a topic you want covered? Send it our way. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

  7. 8月14日

    Anthropic Adds Invisible AI Watermarks to Claude for EU AI Act Compliance

    Hosts Ashley Coffey and Daniel Hill cover two big AI stories this week. First, Anthropic announced that Claude will now embed invisible watermarks in AI-generated text and attach signed "provenance" metadata (a verifiable digital record of where a file came from) to generated files like images, in order to comply with the EU AI Act. Ashley draws a parallel to the EU's USB-C charger mandate to show how EU tech regulations tend to ripple out and affect US companies and consumers too. Second, they dig into Google Gemini's milestone of surpassing 1 billion monthly active users, with a standout stat: 63% of Gemini users talk to it by voice rather than typing — which may explain why Apple chose Gemini to power its revamped Siri.   Links: Anthropic now embeds invisible watermarks in Claude’s text and signed metadata in its files to comply with the EU AI Act. (Claude.com) Google’s Gemini app surges to 1 billion users (TechCrunch)   Connect with Ashley: Instagram Threads LinkedIn   Connect with Daniel: Instagram YouTube Threads   Leave a Review on Apple Podcasts!   Stay curious, stay informed, and stay hydrated! See you next week with more pivotal AI insights. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

  8. 8月11日

    Anthropic Is Building Its Own AI Chips — What It Means for NVIDIA

    Anthropic just confirmed it's building a custom AI chip design team to co-design hardware alongside Claude, because partner GPUs alone can't keep up with demand. Ashley and Daniel break down why every frontier lab is quietly building an exit ramp off general-purpose GPUs, and what that means for NVIDIA's position at the top. Then: Wrinkles, a new iOS and Android app that uses your location to surface the hidden history of the places around you, plus lets families attach their own memories to real places. In this episode: Anthropic's custom silicon play. GPU access has become a bottleneck, not just a cost line. Even with AWS, Google, NVIDIA, and AMD deals in place, Anthropic says renting isn't enough to meet Claude's demand. We cover why co-designing chip and model together cuts inference costs, the same logic behind Google's TPUs, Meta's MTIA, and OpenAI's Broadcom project. What this means for NVIDIA. Daniel's take: if every major AI company builds its own chips, does the most valuable company on the stock market stay that way? Wrinkles, an AI audio tour guide. Free on iOS and Android, it uses your location to surface local history and lore without you staring at your phone. Born out of a frustrating self-guided tour at the British Museum. The family memory angle. Attach photos, videos, and recordings to specific places (a childhood home, a first date, a hometown street) and collaborate with relatives to preserve them. The business model. Free for listeners. Museums, tourism boards, universities, and hotels pay to build location-based experiences instead of funding their own apps. The scale. 1.3 million points of interest across 177 countries, nearly every country on earth. Timestamps (00:04) Welcome and what's on the show (00:40) Anthropic hires a custom AI chip design team (01:07) Why GPU access is a bottleneck, not just a cost (02:13) Daniel's take: the long-term risk for NVIDIA (03:07) Wrinkles, the AI-powered audio tour guide (04:23) Attaching family memories to real places (05:14) How Wrinkles makes money (05:41) Ashley's take: 1.3M points of interest, 177 countries (06:15) Why an Android-at-launch release matters (06:39) Wrap-up and how to support the show Links TechCrunch: Anthropic's custom AI chip team Wrinkles (iOS / Android) Connect Enjoying the show? Leave us a review on Apple Podcasts. It takes a minute, helps others find us, and we'll feature your review on an upcoming episode. Follow Ashley & Daniel on Instagram @ashleyrcoffey89 and @danielhillmedia Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

すべて見る（65）

番組について

Stay informed with AI News You Should Know About, your weekly briefing on the latest AI news and AI technology trends. Join hosts Ashley Coffey & Daniel Hill as they explore key AI headlines, breakthroughs, and the business impact of AI, helping you stay ahead in the ever-evolving tech landscape in 15 minutes or less.

情報