11 NovThe Daily FiveWednesday

Anthropic Prices Below the Hype, and a Court Order Nobody Followed

Anthropic priced its IPO below the rumours, a Delhi court order against deepfakes of the CM got ignored for four days, and India still has no law for the phone-cracking tools its own agencies keep buying. Plus a price-history trick for sale season and whether Spotify's new audiobooks beat Audible.

  1. Long Press · Daily Five11 Nov 202601 / 05
    Illustration for the story: Priced Below The Hype
    ai

    Priced Below The Hype

    Anthropic set its IPO price range at $70 to $78 a share on November 10, targeting a valuation of up to $1.4 trillion, a full trillion below the $2 trillion figure some reports floated just weeks ago. The offering could raise as much as $90.4 billion once you count the $26 billion that Google, Nvidia and five other strategic partners are putting in directly at the IPO price. Trading is expected to start around November 18-19 on the Nasdaq under the ticker ANT, right as a broader AI stock selloff has wiped billions off Nvidia and other AI-linked names this week. The company's revenue run rate hit about $81 billion by late October, up sharply from roughly $71 billion a month earlier, the kind of growth that usually argues for a bigger number, not a smaller one. Buried in the amended filing is a line worth sitting with: Anthropic says recursive self-improvement, AI helping build more capable successor models, could pose a considerable safety risk if it sharply speeds up progress, and that its own systems may be imminently close to that threshold. A company admitting that in its own IPO paperwork, while asking investors for less than the market expected, is choosing to look credible over looking impressive. That's the right call in a market this nervous.

    The question

    What price did Anthropic set for its IPO?

    In one line

    Anthropic set its IPO price range at $70 to $78 a share on November 10, 2026, valuing the company at up to $1.4 trillion and aiming to raise as much as $90.4 billion including $26 billion from partners Google and Nvidia. Trading is expected to start around November 18 to 19 on the Nasdaq under the ticker ANT.

    The problem

    Anthropic needed to go public to fund its compute and growth ambitions, but it was doing so into a market that had already soured on AI valuations, with chip stocks selling off and investors newly wary of how these companies price risk.

    What changed

    On November 10, Anthropic set its IPO range at $70 to $78 a share, implying a valuation of up to $1.4 trillion, well under the roughly $2 trillion number that had circulated weeks earlier. The offering could raise up to $90.4 billion, including $26 billion that Google, Nvidia and five other partners are buying directly, with trading expected to start around November 18-19 under the ticker ANT.

    The catch

    The amended filing also discloses that Anthropic's own systems may be imminently close to the threshold for recursive self-improvement, a risk the company itself flags as potentially serious, and the pricing is happening against an AI stock selloff that has already hit Nvidia, one of Anthropic's own backers.

    What it means for you

    If you're watching AI company valuations as a check on the hype cycle, Anthropic pricing below the rumoured number, while disclosing a real self-improvement risk in writing, is a more honest signal than another record valuation would have been.

    Questions people ask

    How does Anthropic's IPO valuation compare to earlier estimates?

    The $1.4 trillion top-end valuation is well below the roughly $2 trillion figure some reports floated weeks earlier. Anthropic filed confidentially in June, made its prospectus public on October 26, and priced the range on November 10 after an amended filing, with the lower number reflecting a broader AI stock selloff.

    What risk did Anthropic disclose in its IPO filing?

    Anthropic's prospectus warns that recursive self-improvement, where AI helps build more capable successor models, could pose a considerable safety risk if it sharply accelerates AI progress. The company says internal estimates suggest it may imminently meet its Responsible Scaling Policy's threshold for that shift.

    Why are Google and Nvidia buying into Anthropic's IPO?

    Seven strategic partners, including Google and Nvidia, agreed to buy $26 billion of non-voting stock at the IPO price alongside the public offering. Nvidia is a key compute partner for Anthropic, and the stake deepens both companies' financial ties to Claude's maker ahead of its Nasdaq debut.

    SourcesCNBC ↗Fortune ↗

  2. Long Press · Daily Five11 Nov 202602 / 05
    Illustration for the story: Court Said Remove. Still Up.
    newsIndia

    Court Said Remove. Still Up.

    The Delhi High Court ordered six AI-generated videos of Chief Minister Rekha Gupta removed 'forthwith' on November 6, after AAP's posts showed her as a liquor vendor, hawker and delivery agent under a fictional scheme called 'Har Har Rekha, Ghar Ghar Theka.' By November 10, MediaNama found the Facebook video gone but all five flagged X posts still live, one of them with over 512,000 views. Here's the part that stings: India already has a faster route for this. IT Amendment Rules in force since February require platforms to act on flagged synthetic media within two to three hours, yet the order never invokes them, and nobody has explained why a sitting Chief Minister needed a court case for something the rules already promise to handle in an afternoon. Justice Manoj Jain leaned on an earlier Delhi HC ruling involving Raghav Chadha to draw a line between satire and defamation, though MediaNama's own founder argues political satire doesn't stop being satire just because it looks convincing. Either way, the rules and the ruling say the same thing: take it down. Four days on, X hadn't.

    The question

    Are AAP's AI-generated videos of Delhi CM Rekha Gupta still live on X?

    In one line

    Yes. The Delhi High Court ordered six AI-generated videos removed 'forthwith' on November 6, 2026, but MediaNama's check on November 10 found all five flagged X posts, including one with over 512,000 views, still live. Only a Facebook video had been taken down.

    The problem

    AI-generated videos of a sitting Chief Minister circulated widely enough, one clip topped 512,000 views, that she had to sue for defamation to get them removed, even though India's synthetic-media rules are supposed to make takedowns fast and automatic.

    What changed

    The Delhi High Court ordered six AAP posts removed 'forthwith' on November 6. By November 10, the Facebook video was gone, but all five flagged X posts, including the 512,000-view clip, were still live, per MediaNama's check.

    The catch

    India's IT Amendment Rules on synthetic media, in force since February, already require platforms to act on flagged deepfakes within two to three hours. The court order never cites them, and nobody has explained why a case was needed for a takedown the rules already promise.

    What it means for you

    If a platform can sit on a court order about the sitting Chief Minister for four days, don't expect faster action on a deepfake made about you. The legal tools exist; whether they get used before the damage is done is the open question.

    Questions people ask

    What did the Delhi High Court order about the Rekha Gupta deepfake videos?

    Justice Manoj Jain ordered six AI-generated posts removed 'forthwith' on November 6, 2026, after Chief Minister Rekha Gupta sued AAP for defamation. The posts showed her as a liquor vendor, hawker and delivery agent under a fictional scheme called 'Har Har Rekha, Ghar Ghar Theka.' The court issued notices to AAP and other defendants, and the next hearing is December 4.

    Does India have rules that should remove deepfakes faster than a court order?

    Yes. India's IT Amendment Rules on synthetically generated information, in force since February 20, 2026, require significant platforms to label AI content and act on flagged synthetic media within two to three hours for certain complaints. The November 6 court order does not cite these rules, and nobody has explained why they weren't used first.

    How does this case compare to the Raghav Chadha deepfake ruling?

    In July 2026, the Delhi High Court refused a blanket takedown of AI content about Raghav Chadha, treating most of it as political satire, but ordered six explicit items removed. Justice Jain cited that case while ruling that AAP's content about Gupta crossed from satire into defamation of her personal dignity.

    SourcesBar and Bench ↗MediaNama ↗

  3. Long Press · Daily Five11 Nov 202603 / 05
    Illustration for the story: That Discount Might Be Fake
    tip

    That Discount Might Be Fake

    Every big sale season, the same trick shows up: a product's 'original price' quietly climbs right before the discount badge appears, so a modest markdown reads like a huge one. Amazon's in-app Price History feature is built to catch that. Open any product page, tap Price History, and you get up to 365 days of what the item has actually sold for, not what today's banner claims. Don't want to dig through the page? Just ask Rufus, Amazon's built-in AI assistant, 'has this price dropped before,' and it tells you directly. It isn't bulletproof. Some sellers still rotate listings or tweak variants to dodge a clean price trail, so a chart with no real history isn't proof either way. But for most products, it turns 'biggest sale of the year' from a slogan into a number you can check before you add to cart.

    The question

    How can I tell if an Amazon India discount is real?

    In one line

    Open the product page on the Amazon app and tap Price History to see up to 365 days of what an item has actually cost, or ask Rufus, Amazon's AI shopping assistant, 'has this price dropped before.' Both show whether today's 'sale' price is genuinely lower than recent prices or just a reset 'original' price.

    The problem

    Big sale events train shoppers to trust a percentage-off badge, even though retailers can inflate the 'original price' right before a discount to make an ordinary markdown look dramatic.

    What changed

    Amazon's app has quietly built in a Price History view that shows up to 365 days of a product's actual price, reachable from the product page or by asking the Rufus AI assistant directly whether a price has dropped before.

    The catch

    The feature only works as well as Amazon's own price-tracking data for that listing, so very new products or ones sellers frequently relist can show incomplete history, and it doesn't track third-party seller pricing on the same listing.

    What it means for you

    Before trusting any 'X% off' badge on Amazon India, especially during a big sale, tap Price History or ask Rufus if it has dropped before. It takes ten seconds and tells you whether you're getting a real deal or just reset math.

    Questions people ask

    Where do I find Amazon's Price History feature?

    On a product's listing page in the Amazon app, look for the Price History option below the price, which plots up to a year of past prices. If you don't see it, Amazon's Rufus chat assistant can answer the same question directly when you type or ask 'has this price dropped before.'

    Does Price History work for every product on Amazon India?

    No. The feature depends on Amazon having tracked that specific listing's price over time, so very new listings or ones that change ASINs frequently may show limited or no history. It also only reflects Amazon's own listed price, not third-party seller price changes on the same listing.

    Why do retailers inflate the 'original price' before a sale?

    A higher struck-through 'original price' makes the sale discount look bigger than it is, a common tactic during big sale events. Checking a verified price history, rather than trusting the percentage-off badge, is the simplest way to confirm whether a deal is genuinely new or just restated math.

  4. Long Press · Daily Five11 Nov 202604 / 05
    Illustration for the story: Spotify vs Audible: India
    comparisonIndia

    Spotify vs Audible: India

    Spotify's audiobooks land in India on November 12 through Audiobooks+, an add-on for Premium Family and Duo subscribers that unlocks 15 hours of listening a month from a catalog of more than 350,000 titles in 120-plus languages. It's part of a bigger global push, Spotify is expanding audiobooks from 22 markets to more than 180 by the end of the year, but the company hasn't said what the add-on will actually cost in India. Fifteen hours sounds generous until you remember most audiobooks run eight to fifteen hours themselves, so that cap is roughly one book a month for an average listener. Audible has had an answer sitting in India since 2018: Rs 199 a month, one credit to buy and keep an audiobook forever, plus unlimited streaming of its Originals and Audible Plus catalog with no credit needed. One is a rental cap with an unknown price tag. The other is a flat fee you already know, with a book that's actually yours at the end of it. Until Spotify names a number, there isn't really a contest.

    The question

    Is Spotify's Audiobooks+ or Audible better value in India?

    In one line

    Spotify brings Audiobooks+ to India on November 12, 2026, giving Premium Family and Duo subscribers 15 hours of audiobook listening a month from a catalog of over 350,000 titles, with no India price announced for the add-on yet. Audible India has offered Rs 199 a month since 2018 for one credit to keep an audiobook forever plus unlimited streaming of its Originals catalog, which makes it the clearer deal for now.

    The problem

    Indian listeners who want audiobooks inside an app they already use have had to choose between a dedicated audiobook app and whatever their music streaming service offers, and Spotify is only now closing that gap.

    What changed

    Spotify is launching Audiobooks+ in India on November 12, 2026, giving Premium Family and Duo subscribers 15 hours of monthly audiobook listening from a 350,000-plus title catalog, as part of expanding the feature from 22 to more than 180 markets worldwide.

    The catch

    Spotify has not disclosed what Audiobooks+ will cost as a standalone add-on in India, and 15 hours a month covers roughly one average-length audiobook, while Audible's existing Rs 199-a-month plan gives a book to keep forever plus unlimited Originals streaming.

    What it means for you

    If you listen to more than one audiobook a month, Audible's flat Rs 199 with a book to keep still beats an unpriced 15-hour cap. Wait for Spotify to actually name an India price before switching anything.

    Questions people ask

    What does Spotify's Audiobooks+ include in India?

    Audiobooks+ is an add-on for Spotify Premium Family and Duo plans that gives 15 hours of audiobook listening a month from a catalog of more than 350,000 titles across 120-plus languages. It launches in India on November 12, 2026, as part of Spotify's expansion from 22 to more than 180 markets, but Spotify hasn't disclosed what it will cost in India.

    How much does Audible cost in India?

    Audible India has charged Rs 199 a month since launching in the country in 2018. Subscribers get one credit a month to buy and permanently keep an audiobook, plus unlimited streaming access to Audible's Originals and Audible Plus catalog, which doesn't require using a credit.

    Is 15 hours a month enough for a regular audiobook listener?

    It depends on the book, but most full-length audiobooks run 8 to 15 hours, so a 15-hour monthly cap on Spotify's add-on covers roughly one average title a month. Audible's model gives unlimited streaming of its Originals catalog on top of one title to keep, which suits a heavier listening habit better.

    SourcesT2 Online ↗91mobiles ↗

  5. Long Press · Daily Five11 Nov 202605 / 05
    Illustration for the story: No Law. Just A Tool.
    myth busterIndia

    No Law. Just A Tool.

    There is no law in India that says when the police can crack open your phone. Not a statute, not a published rule, nothing that specifies what can be copied off the device, how long the data sits in a file somewhere, or who gets to read it. The Internet Freedom Foundation dug this up through an RTI request: Home Minister Amit Shah personally signed off on a one-time exemption from Make in India procurement rules so the National Investigation Agency could buy a Cellebrite Premium Tool, a three-year licence running about Rs 4.98 crore, awarded to a company called Nextechno Gen. The tool's own specification calls for brute-forcing passcodes across most Android and iPhone models and pulling a device's entire file system, including protected areas like Samsung's Secure Folder. Cellebrite's tools have shown up in Serbia, Jordan, Kenya and Russia, used against journalists and activists, according to researchers who've tracked them. Fifteen separate Cellebrite-related tenders appear across Indian agencies this year alone, from the Serious Fraud Investigation Office to state forensic departments. And then there's Dhruv Rathee, detained by Delhi Police on October 10 ahead of a protest, his phone and his parents' phones taken. He says officers may have tried to image his iPhone. Unverified, by IFF's own account. What isn't unverified is that the tool exists, the money's been spent, and nobody can point you to the rule that's supposed to govern any of it.

    The question

    Is there a law governing when Indian police can unlock someone's phone?

    In one line

    No. India has no statute, rule, or published protocol specifying when a phone can be unlocked, what can be copied from it, how long the data is kept, or who can access it. The Internet Freedom Foundation's RTI findings show Home Minister Amit Shah personally approved a Make in India exemption so the NIA could buy a Cellebrite Premium phone-cracking tool for about Rs 4.98 crore.

    The problem

    Indian law enforcement agencies are acquiring powerful phone-extraction tools at scale, with at least 15 Cellebrite-related tenders across agencies this year, while no law specifies the conditions under which someone's phone can be unlocked, copied, or how long that data is retained.

    What changed

    The Internet Freedom Foundation's RTI findings show Home Minister Amit Shah personally approved a Make in India exemption so the NIA could buy a three-year Cellebrite Premium Tool licence for about Rs 4.98 crore, a tool built to brute-force passcodes and extract full file systems from Android and iPhone devices.

    The catch

    No Indian statute governs what can be copied from an unlocked phone, how long it's retained, or who can access it, and the claim that Delhi Police tried to image Dhruv Rathee's iPhone after detaining him in October remains unverified, with the police not responding publicly.

    What it means for you

    The capability to fully extract anyone's phone exists and is spreading across Indian agencies faster than any law to govern it. That gap matters regardless of which side of a protest, a newsroom, or a court case you happen to be on.

    Questions people ask

    What did Amit Shah approve for the NIA?

    A note dated January 17, 2026 shows Home Minister Amit Shah approved a one-time exemption from Make in India procurement preference rules, letting the National Investigation Agency buy a Cellebrite Premium Tool with a three-year licence for about Rs 4.98 crore, awarded to Nextechno Gen Private Limited.

    What can Cellebrite's tools actually do to a phone?

    Cellebrite's Premium tool specification calls for brute-forcing passcodes on many Android and iPhone models, extracting a device's full file system, and pulling data from protected areas such as Samsung's Secure Folder. Forensic researchers have documented similar Cellebrite tools being used against journalists and activists in countries including Serbia, Jordan, Kenya and Russia.

    What happened to Dhruv Rathee's phone?

    YouTuber Dhruv Rathee was detained by Delhi Police on October 10, 2026, ahead of a planned protest, and his phone along with his parents' phones were taken. Rathee has said police may have tried to image his iPhone using Cellebrite, though the Internet Freedom Foundation notes this specific claim is unverified and Delhi Police hasn't responded publicly.

    SourcesInternet Freedom Foundation ↗

Taps

Quick hits. One line each, for when a long press is too much.

  • India

    MakeMyTrip buys majority stake in Quick Ride, enters ride-pooling business

    The ride-sharing app, whose enterprise clients include TCS and Infosys, moves into a new entity, Co Ride Mobility, under MakeMyTrip. Inc42

    Inc42 ↗
  • India

    Kerala HC lets NoCap resume its Facebook page after state denies seeking block

    Meta had restricted the Malayalam news page over a notice never made public; the state says it only flagged hateful comments, not the whole page. The News Minute

    The News Minute ↗
  • Global

    Thailand's SEC to block crypto exchange Bitget from December 5 over unlicensed trading

    Regulators allege Bitget ran a Thai-language, baht-priced exchange without a licence from 2022 to 2025; users are told to close positions before the block. CoinGape

    CoinGape ↗
  • Global

    Windows 11's November update adds a faster right-click menu and quicker Edge launches

    KB5132061 also adds a USB-free cloud rebuild for broken installs and claims up to 20% faster Start, Search and File Explorer on 8GB PCs. Windows Latest

    Windows Latest ↗
  • Global

    NASA unveils X-70 Fractal, a crewed research jet eyeing first flight by 2028

    The 10,000-lb aircraft carries up to 2,000 lb of swappable experimental payloads and is meant to flight-test new control tech in real air, not simulators. NASA

    NASA ↗
That's all five. Next edition tomorrow.Press and hold