Opus 5.5 is not better than 6.1 sol in my workflow
With better uncertainty estimates and a joint distribution over all models’ ECIs, we can now compare models directly.
For example, we can now say Astra has a 72% probability of being more capable than Opus 5.5, but a 99% probability of being more capable than Fable 5.1.
ALT Dot-and-whisker chart of the ECI gap between GPT-6 Astra and each of the five next-highest models under the new method, with 90% intervals. The probability that GPT-6 Astra is more capable is 54% against GPT-6.1 Sol, 72% against Claude Opus 5.5, 99% against Claude Fable 5.1 and Claude Sonnet 5.5, and 100% against Gemini 4 Argon.
kai retweeted
That’s how Wenfeng’s “true American frontier is 800B active” looks, I suppose. Plus some looping probably.
kai retweeted
Oracle cancels $7 billion chip leasing deal with Tencent — FT
#MacroEdge
kai retweeted
US OFFICIALS HAVE HELD UP APPROVAL FOR AMERICAN CHIPMAKING EQUIPMENT NEEDED BY SK HYNIX TO EXPAND PRODUCTION IN CHINA - PEOPLE FAMILIAR SAY
kai retweeted
"In the race to win AI builders, Anthropic and OpenAI now account for 86% of enterprise LLM API spend, up from 67% last year. As the market consolidates around the frontier leaders, model capability—not price—is proving to be the decisive competitive advantage." - Menlo Ventures
Meanwhile, Google's market share crumbled - down more than 50%! Data below:
More bad news for Google:
“Right now, the best models are capturing the lion’s share of enterprise,” said Deedy Das, Partner at Menlo Ventures. “Our data shows that frontier labs capture nearly seven of every ten dollars spent on enterprise AI, up from half last year. Anthropic grew its share of enterprise API spending from 40% to 55%, while Google fell from 21% to 9%. That’s how quickly the competitive landscape can change."
kai retweeted
The heads of Marvell Technology and Lumentum Holdings got the blessing of Nvidia’s Jensen Huang before striking their nearly $103 billion tie-up. bloomberg.com/news/articles/…
kai retweeted
$AEHR TEST SYSTEMS Q1’27 EARNINGS HIGHLIGHTS
🔹 Revenue: $33.0M (Est. $27.1M) 🟢; +200% YoY
🔹 Adj. EPS: $0.28 (Est. $0.11) 🟢; +2700% YoY
🔹 Bookings: $46.7M
🔹 Effective Backlog: $109.3M
Raises FY27 Guide:
🔹 Revenue: $150M (Est. $137M) 🟢
🔹 Non-GAAP Net Income: At least 20% of revenue
🔹 Prior Revenue Guide: $130M-$150M
Other Q1 Metrics:
🔹 Backlog: $100.9M
🔹 GAAP EPS: $0.21
🔹 Cash: $105.1M
Key Updates:
🔹 AI Processor WLBI: $22M follow-on order announced in August 2026
🔹 Sonoma: More than a dozen systems shipped and installed per month
🔹 Customer Base: Potential across more than a dozen customers
kai retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
kai retweeted
In a weird way, everybody is rooting for you. When you go to a job interview, the interviewer wants you to be amazing so they don’t have to look anymore. when you go on a date, the person wants to have an amazing date with you. If you make a piece of content, people want to see amazing pieces of content and they have no problem if it’s yours.
the instant you get into the mindset that people want you to do well, your entire world gets better. And you’ll be shocked to know this pattern keeps repeating in different areas of life.
kai retweeted
I asked Claude: "What types of businesses outside of the labs will get stronger as raw intelligence explodes?"
Honestly, spot on response.
kai retweeted
Anthropic says its internal estimates and expectations for future models suggest it may soon reach its threshold for dramatically accelerated AI progress driven by AI itself.
kai retweeted
Amended S-1/A is out
cnbc.com/2026/11/02/anthropi…
Looks like @FundaAI was correct, more interestingly only a portion of this $5B additional xAI capacity is used for inference
FYI @stevehou
kai retweeted
human-in-the-loop
└ human-on-the-loop
└ human-near-the-loop
└ human-cc'd-on-the-loop
└ human-legally-liable-for-the-loop
kai retweeted
ANTHROPIC Q3 2026 FINANCIAL HIGHLIGHTS:
- Revenue: $17.0B, +1,127% YoY & 47% QoQ
- Oper Loss: $567M (vs. $974M profit in Q2)
- Net Loss: $23.3B, vs. $12.1B YoY
- Adj. Oper Income: $2.5B
- Adj. Oper Margin: 15%, down from 19% in Q2
- 9M Revenue: $33.2B
Anthropic attributed its return to an operating loss partly to $5B in higher-cost short-term compute and a $2B noncash donation-matching charge.
Its nine-month net loss reached $104B, largely reflecting $94.3B in noncash charges from the rising value of its convertible notes.
Anthropic also disclosed approximately $895.4B in aggregate purchase and lease commitments as of and after Sept. 30, largely tied to AI compute infrastructure.
