frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Anthropic may not secure investment-grade rating after IPO, Blackstone says

https://www.ft.com/content/59383c7f-8d82-493f-bf01-d2e862a1018f
71•petethomas•1h ago

Comments

rzz3•1h ago
Paywall; I wish HN would just ban paywalled posts.
youngtaff•58m ago
What's wrong with paying for serious news?
radlad•54m ago
Nothing wrong with paying for serious news, but when most viewers can't read the article it probably won't get much useful discussion.
furyofantares•54m ago
The problem is sharing it to a group of people who mostly aren't paying for it. It's a good way to get a thread that's just people reacting to the title, which drives HN toward just being a discussion board instead of a place to find good content.
sreekanth67•48m ago
I would pay if its token based pay per read model. nobody want to subscribe to 10 news websites paying hundreds every month.
wslh•7m ago
This, don't get why there was no model like this. Even understanding the complexity to reach consensus (and technology).
eqvinox•46m ago
Agree with the 2 siblings: the problem is HN readers at large unable to read it, hence no useful discussion.

I'd also add: from paid online news, I also expect a moderated forum/comment section, with a human to escalate to if automatic/AI moderation fails.

isatty•40m ago
1. Can’t be bothered

2. Never going to visit this site again and I am not going to make an account for it

I am however, okay with turning off the ad blocking if they’re running reasonable ads (not every other paragraph).

paxy•18m ago
Should everyone pay for every news source?
lurk2•16m ago
People who participate in forums share their insights for free and resent the friction that a paywall adds to information being diffused.
radlad•59m ago
https://wallhop.io/https://www.ft.com/content/59383c7f-8d82-...
mosesman•58m ago
https://giftarticle.ft.com/giftarticle/actions/redeem/2d0b4d...

9 shares left

please comment if you have used it to ensure other people know if it could be read

and its limited - only click if you are interested

enjoy!

daviTeodoro•53m ago
He is the article: Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

Blackstone told investors Anthropic may not receive an investment-grade rating immediately after going public, a reversal from its previous guidance that would complicate the AI lab’s ability to finance its massive spending.

The private capital giant told investors in Anthropic’s latest $63bn chip financing deal that the start-up is expected to receive a “high” junk rating from major credit rating agencies after its initial public offering, according to people familiar with the matter.

Blackstone is acting as an anchor investor on the deal, known as “Project North Star”, and is working with banks that stumped up financing on selling the debt on to other investors.

Anthropic filed to go public on Monday, setting up what could be the largest listing of all time next month. The AI lab’s prospectus sketched out its “bold but sane” plans to spend more than $800bn on computing infrastructure between now and 2045.

An investment-grade credit rating is central to those ambitions, enabling Anthropic to borrow at lower rates to finance its AI infrastructure projects. The group has cash and equivalents of more than $112bn, and investors anticipate it could raise about $100bn more in the IPO. Sign reading "Blackstone" outside the Blackstone headquarters in New York, with pedestrians and city lights reflected in the glass. Blackstone’s funds have invested in $9bn of the junior debt at a lucrative yield of more than 10% © Angus Mordant/Bloomberg

While Blackstone had told its investment partners in recent months that it was expecting an investment-grade rating for Anthropic before year-end, it has recently signalled that a rating in the double B range or above is more likely given the guidance it has received from major credit agencies, the people said.

Such a designation would limit the pool of capital that Anthropic could tap because many insurance and pension funds are restricted from taking large positions in junk-rated borrowers. As a result, the company might need to continue to rely on its suppliers with high credit ratings to support its debt-raising efforts.

US tech stocks slid to session lows on Friday following the FT’s report on Anthropic’s rating, with the Nasdaq 100 down 3.7 per cent in afternoon trading. Markets were sharply lower prior to the report after earnings from Amazon and Intel worsened fears about the sustainability of the AI boom.

“It’s all about access to credit, it really matters . . . if [Anthropic] isn’t investment grade, the cost of credit goes higher, that’s not good for anyone,” said a US tech-focused portfolio manager.

“It’s a bloodbath out there,” they added, referring to the sell-off in stocks.

Wall Street banks earlier this week closed the North Star deal, which provides the financing for Anthropic to lease 1.5 gigawatts of Google’s semiconductors. More than $42bn of the senior loans is guaranteed by Broadcom, which co-develops the chips and has an investment-grade rating, while $18bn of the junior debt is backed by the AI start-up’s lease payments.

Institutional investors covet senior debt because it takes priority over junior loans in the case of a default. Broadcom’s high A-tier credit rating was an important factor in allowing Anthropic to secure the financing.

Blackstone’s funds invested in $9bn of the junior debt at a lucrative yield of more than 10 per cent, the FT previously reported. Bankers for Anthropic and rival OpenAI have also lobbied the largest rating agencies to grant the AI labs investment-grade ratings after their IPOs.

Project North Star’s $18bn junior debt, which exposes lenders more acutely to Anthropic’s credit risk, has been assigned a double B plus rating with positive outlook from Kroll Bond Rating Agency, a slight improvement from stable outlook in a similar deal known as Project “Big Sky” four months ago. It also received a double B rating from Morningstar DBRS, the people added.

The big three credit agencies — Fitch, Moody’s and S&P — which have bigger influence on companies’ borrowing costs, have not officially issued any credit ratings for Anthropic.

Anthropic, Blackstone, KBRA, Morningstar DBRS and S&P declined to comment. Fitch and Moody’s did not respond to requests for comment.

_HMCB_•51m ago
What could go wrong?
blocked_againnn•10m ago
Guys guys but AI is conscious.........

And it's going to kill us us all.

Please buy our stocks please..

Machines of love and grace.

We are building it.

It's going to build a great earth. Like kingdom of God.

Buy stocks. Please.

Ohh and we are making a wet lab..

It's going to be great. Buy our stocks please.

Don't say our stock is junk.

It would hurt Claude.

Just buy our stock instead.

Or roko basilisk will kill us all.

cmiles8•5m ago
Is this a surprise?

By any sane normal sense of business and financials these AI labs look terrible on paper.

Any loan would be speculative at best, and the rating would simply reflect that.

jacquesm•24m ago
"bold but sane" -> that sounds as if Claude wrote their prospectus?
neutronicus•46m ago
> Wall Street banks earlier this week closed the North Star deal, which provides the financing for Anthropic to lease 1.5 gigawatts of Google’s semiconductors.

This line is certainly eyebrow-raising to the educated layman.

Apparently it means that anthropic is leasing a Google data center whose constituent servers draw 1.5 GW of power running representative workloads.

joe_the_user•20m ago
It's interesting you imagine nine shares would suffice for a large site like HN. No, the link doesn't work for me. Honestly, this kind of "sharing" seems a bit disingenuous at this point.
hoppp•18m ago
Yeah, 9 shares and then 10k people click

My favourite PowerMac is a Nintendo Wii

https://virtuallyfun.com/2026/10/29/my-favourite-powermac-is-a-nintendo-wii/
82•DHowett•1h ago•12 comments

GitHub Incident with Actions, Pages, and Webhooks

https://www.githubstatus.com/incidents/pwxsjgnvrk6b
66•damontal•1h ago•26 comments

Northflank Breached

https://status.northflank.com
208•js4ever•3h ago•130 comments

Backdoor in Claude Code for VS Code extension on the Cursor Marketplace

https://phorion.io/blog/offical-claude-code-vsx-compromise/
515•_chall•11h ago•394 comments

A Working Nintendo in ~1GB of Qwen Weights

https://github.com/dbreunig/nesformer
94•dbreunig•2h ago•81 comments

Florilegium: Procedural Flowers in LaTeX

https://github.com/kevinraymond/florilegium
87•nanna•3h ago•17 comments

The federated code forge

https://tangled.org/blog/forge/
117•zlliang•3h ago•42 comments

Breaking the open web won't stop the bad bots

https://www.techpolicy.press/breaking-the-open-web-wont-stop-the-bad-bots/
55•raybb•3d ago•18 comments

AI isn't replacing SREs, it's deskilling them

https://newsletter.signoz.io/p/ai-isnt-replacing-sres-its-deskilling
322•dorianniemiec•3h ago•239 comments

Azure's Weakest Link – Five Full Cross-Tenant Compromises

https://www.binarysecurity.no/posts/2026/10/one-root-case
210•swq115•9h ago•36 comments

Anthropic may not secure investment-grade rating after IPO, Blackstone says

https://www.ft.com/content/59383c7f-8d82-493f-bf01-d2e862a1018f
75•petethomas•1h ago•21 comments

Blockchain intelligence firms link North Korea to $6.3B hack on Binance

https://www.npr.org/2026/10/30/g-s1-150902/blockchain-intelligence-firms-link-north-korea-to-6-3b...
120•paulpauper•3h ago•68 comments

How do you take a picture of a black hole?

https://www.quantamagazine.org/how-do-you-take-a-picture-of-a-black-hole-20261029/
62•dverlaeckt80•9h ago•45 comments

Building an AI-native engineering org

https://twitter.com/sgondala2/status/2115832070274723875
17•sgondala_ycapp•21h ago•16 comments

Show HN: Spinf-serve – inference server for bulk scoring and decisions

https://github.com/spinfinc/spinf-serve
8•giarec•2h ago•0 comments

Show HN: Sketch Stadium – watch your agent play Pictionary

https://sketch.wakeshift.ai/
6•ericlevine•3h ago•0 comments

French national assembly votes 0.055€/GB internet tax

https://www.lesnumeriques.com/societe-numerique/netflix-steam-cloud-vos-abonnements-et-telecharge...
23•nick486•35m ago•19 comments

Tracks along the longest drive on another world

https://science.nasa.gov/blog/tracks-along-the-longest-drive-on-another-world/
37•shcallaway•17h ago•2 comments

Show HN: I built a multiplayer game where you're an aerial fighter cat

https://catcopter.com
42•pawelwentpawel•3h ago•20 comments

Compile-Time Dependent Types

https://thenumb.at/Syl-Intro/
42•ibobev•4d ago•11 comments

Executable Archaeology: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984)

https://arxiv.org/abs/2610.36341
22•abrax3141•2d ago•0 comments

The $6.3B Binance hack was pulled off in just 20 minutes

https://www.wired.com/story/the-dollar63-billion-binance-hack-was-pulled-off-in-just-20-minutes/
246•domhudson•6h ago•174 comments

Why a program doesn't work when run from a batch file, but works when clicked

https://devblogs.microsoft.com/oldnewthing/20261026-00/?p=112797/
106•ibobev•3d ago•58 comments

Polymarket 1h halt after vendor compromise; affected users will be 'made whole'

https://www.nbcnews.com/business/consumer/polymarket-trading-halt-hack-wallets-rcna606405
77•theanonymousone•3h ago•23 comments

Gilgamesh by Simon Armitage review – an ancient epic is brilliantly translated

https://www.theguardian.com/books/2026/oct/20/gilgamesh-by-simon-armitage-review-an-ancient-epic-...
166•benbreen•3d ago•51 comments

Finding Panics with Call Graph Analyzer

https://tweedegolf.nl/en/blog/242/finding-panics-with-call-graph-analyzer/
59•ndr•2d ago•3 comments

Pell Puzzles Review

https://dles.gg/reviews/pell-puzzles
15•trizoza•2d ago•3 comments

Medicare Will Pay $50k for Your Heart Attack. It Should Pay $100 to Prevent It

https://www.medpagetoday.com/opinion/second-opinions/123553?trw=no
225•brandonb•4h ago•217 comments

Wasmer (YC S19) Is Hiring OS/Hypervisor Experts in Europe

https://www.workatastartup.com/jobs/98722
1•syrusakbary•12h ago

Proton's CEO Cancels AI 'Re-Interviews'

https://www.404media.co/protons-ceo-cancels-ai-re-interviews/
173•cdrnsf•1h ago•136 comments