Patrick Moorhead

69.8K posts

Patrick Moorhead banner
Patrick Moorhead

Patrick Moorhead

@PatrickMoorhead

Founder, CEO, Chief Analyst @MoorInsStrat. Co-founder of @TheSixFiveMedia and @Signal_65. Healthspan improver. Ex-AMD CVP, Compaq. Use AI to improve posts.

Austin, TX Katılım Mayıs 2008
4.6K Takip Edilen60.9K Takipçiler
Sabitlenmiş Tweet
Patrick Moorhead
Patrick Moorhead@PatrickMoorhead·
The Week Ahead In Technology for April 13, 2026. $TSM and $ASML report: the two biggest reads on AI semiconductor demand. Section 232 tariff deadline could reshape chip imports. @OpenAI vs @AnthropicAI rivalry escalates on revenue and compute.
Patrick Moorhead@PatrickMoorhead

The Week Ahead In Technology for April 6: HumanX opens at Moscone this week with 6,000+ attendees. AWS's CEO Garman, OpenAI's B2B CTO Narayanan, @Anthropic’s Krieger, and @drfeifei headlining. The conversation has shifted from capability to deployment and ROI. That is the signal. $TSM March revenue drops this week– the AI demand read ahead of Q1 earnings. ~70% global foundry share means this number sets the tone. $INTC ’s $14.2B Fab 34 buyback from Apollo. $ORCL cutting thousands to fund $50B AI infra. $SNX record Q1 beat with all five top U.S. hyperscalers now signed. Enterprise AI’s infrastructure moment is here. The question is who captures the ROI.

English
0
1
11
4.6K
Peter Yang
Peter Yang@petergyang·
My entire feed and the Claude subreddit is full of ppl saying opus got nerfed. Why would Anthropic nerf its own models?
English
298
11
663
119K
Patrick Moorhead
Patrick Moorhead@PatrickMoorhead·
This is the best assessment I’ve read so far on what’s happening in the enterprise with AI and agents. It mirrors many of the conversations the team and I are having. One thing Aaron didn’t address directly here (maybe inferred) was the desire for a central orchestrator, one that manages many of the agents across the enterprise. Cloud companies are building inside their sandboxes but the reality is that every enterprise is heterogeneous and cloud agents don’t run or manage on prem agents. AI companies manage their own agents even if they do span location but don’t want to be locked into one vendor. Aaron addressed this for agents but not orchestration. I spent hours with the former CIO of one of the largest institution and she told me that the enterprise is lacking even enough people to plan, manage, and deploy many projects at the same time even if the agents are doing most of the heavy lifting. It’s important we separate technological capability with the enterprises ability to manage and deploy it. I see next gen models as helping but not changing the fundamental enterprise issue.
Aaron Levie@levie

Another week on the road meeting with a couple dozen IT and AI leaders from large enterprises across banking, media, retail, healthcare, consulting, tech, and sports, to discuss agents in the enterprise. Some quick takeaways: * Clear that we’re moving from chat era of AI to agents that use tools, process data, and start to execute real work in the enterprise. Complementing this, enterprises are often evolving from “let a thousand flowers bloom” approach to adoption to targeted automation efforts applied to specific areas of work and workflow. * Change management still will remain one of the biggest topics for enterprises. Most workflows aren’t setup to just drop agents directly in, and enterprises will need a ton of help to drive these efforts (both internally and from partners). One company has a head of AI in every business unit that roles up to a central team, just to keep all the functions coordinated. * Tokenmaxxing! Most companies operate with very strict OpEx budgets get locked in for the year ahead, so they’re going through very real trade-off discussions right now on how to budget for tokens. One company recently had an idea for a “shark tank” style way of pitching for compute budget. Others are trying to figure out how to ration compute to the best use-cases internally through some hierarchy of needs (my words not theirs). * Fixing fragmented and legacy systems remain a huge priority right now. Most enterprises are dealing with decades of either on-prem systems or systems they moved to the cloud but that still haven’t been modernized in any meaningful way. This means agents can’t easily tap into these data sources in a unified way yet, so companies are focused on how they modernize these. * Most companies are *not* talking about replacing jobs due to agents. The major use-cases for agents are things that the company wasn’t able to do before or couldn’t prioritize. Software upgrades, automating back office processes that were constraining other workflows, processing large amounts of documents to get new business or client insights, and so on. More emphasis on ways to make money vs. cut costs. * Headless software dominated my conversations. Enterprises need to be able to ensure all of their software works across any set of agents they choose. They will kick out vendors that don’t make this technically or economically easy. * Clear sense that it can be hard to standardize on anything right now given how fast things are moving. Blessing and a curse of the innovation curve right now - no one wants to get stuck in a paradigm that locks them into the wrong architecture. One other result of this is that companies realize they’re in a multi-agent world, which means that interoperability becomes paramount across systems. * Unanimous sense that everyone is working more than ever before. AI is not causing anyone to do less work right now, and similar to Silicon Valley people feel their teams are the busiest they’ve ever been. One final meta observation not called out explicitly. It seems that despite Silicon Valley’s sense that AI has made hard things easy, the most powerful ways to use agents is more “technical” than prior eras of software. Skills, MCP, CLIs, etc. may be simple concepts for tech, but in the real world these are all esoteric concepts that will require technical people to help bring to life in the enterprise. This both means diffusion will take real work and time, but also everyone’s estimation of engineering jobs is totally off. Engineers may not be “writing” software, but they will certainly be the ones to setup and operate the systems that actually automate most work in the enterprise.

English
0
0
0
1.4K
Stacy Rasgon
Stacy Rasgon@Srasgon·
Poor doggo wearing the cone of shame
Stacy Rasgon tweet media
English
7
0
29
10.6K
Patrick Moorhead
Patrick Moorhead@PatrickMoorhead·
@TerryLennox_ @alexeheath They’ve run out of compute and are dialing down every knob they can to continue. Everything but raw CLI API from them gets throttled. There you can choose four different Opus strengths. Chat, Cowork and Code app are all dumb and throttled.
English
0
0
1
38
Alex Heath
Alex Heath@alexeheath·
Just me or has Opus been majorly nerfed in last couple days? On a Max plan
English
32
4
179
15.5K
Patrick Moorhead
Patrick Moorhead@PatrickMoorhead·
@Baxate It took me too long to figure this out. I’m glad you see this.
English
0
0
1
250
Baxate
Baxate@Baxate·
if you say you are serious about improving yourself/ your life and you don’t prioritize getting good sleep consistently you are a liar
English
4
3
46
1.8K
Apoorv Agrawal
Apoorv Agrawal@apoorv03·
MS&E 435, Week 2. Thank you @altcap and @sundeep for giving students the real playbook w no filters. And to the students who packed this room and asked questions that made even the guests think: you’re the reason this class works. Can’t wait for Week 3!
Apoorv Agrawal tweet media
English
9
6
111
9.3K
Patrick Moorhead
Patrick Moorhead@PatrickMoorhead·
Is this real? Could Opus hallucinations possibly have gone from #2 to #12? I have experienced it. It’s good to see real numbers documenting the experience.
BridgeMind@bridgemindai

CLAUDE OPUS 4.6 IS NERFED. BridgeBench just proved it. Last week Claude Opus 4.6 ranked #2 on the Hallucination benchmark with an accuracy of 83.3%. Today Claude Opus 4.6 was retested and it fell to #10 on the leaderboard with an accuracy of only 68.3%. A 98% increase in hallucination. bridgebench.ai just confirmed that Claude Opus 4.6 has reduced reasoning levels and is nerfed.

English
8
2
27
4.3K
Sharon Goldman
Sharon Goldman@sharongoldman·
Anthropic's Claude Code isn't ready for the rest of us. Not even close. Developers told me I could get going in just a few minutes. They were right — but only with my developer husband by my side. sharongoldman.substack.com/p/anthropics-c…
English
2
4
16
1.8K
Patrick Moorhead
Patrick Moorhead@PatrickMoorhead·
What a run.
Amazon@amazon

.@awscloud launched its first custom cloud processor in 2018. Five generations later, Graviton chips have become the backbone of modern cloud infrastructure. Traditional server chips weren't built for the cloud. That's where Amazon saw an opportunity and invented Graviton, a family of custom chips designed specifically for cloud workloads—making applications faster, cheaper, and more energy efficient. Fast forward to today: 90,000+ customers run on Graviton, including 98% of AWS's top 1,000 customers. Better performance at a lower cost. ⚡

English
1
4
41
10.4K
DogeDesigner
DogeDesigner@cb_doge·
Anthropic’s Claude Opus is FALLING. Latest benchmarks show its accuracy dropped from 83.3% → 68.3% in just days. That’s a major spike in hallucinations during coding. Grok 4.20 still holds the #1 spot. Undefeated.
DogeDesigner tweet media
English
216
169
1.8K
104.3K
Patrick Moorhead
Patrick Moorhead@PatrickMoorhead·
@dee_bosa I have an agent that takes my raw Week In Technology video, adds subtitles, highlights key words, adds logos, b-roll and salutation page.
English
0
0
1
390
Deirdre Bosa
Deirdre Bosa@dee_bosa·
the software disruption I've been covering just showed up in my own workflow In Feb I vibecoded a tool to cut clips from my videos. Took 27 minutes. Couldn't do subtitles. On Friday (2 months later): same task took seconds
Deirdre Bosa tweet mediaDeirdre Bosa tweet media
English
11
8
73
7.3K
Anush Elangovan
Anush Elangovan@AnushElangovan·
What agents do to your sleep schedule:
Anush Elangovan tweet media
English
3
1
28
3.1K
Patrick Moorhead
Patrick Moorhead@PatrickMoorhead·
The pod is up! We normally post on Monday mornings, but are testing publishing it on Friday-Saturday.
Six Five Media@TheSixFiveMedia

Ep. 300 of The Six Five Pod hits the real pressure points in AI right now. Anthropic unveiled Mythos and held it back. OpenAI signaled a major leap with Spud. Meta launched Muse Spark. Intel stacked moves across Terafab, packaging, and inference. Maine lawmakers moved toward a data center freeze. Google, Broadcom, and Anthropic locked in compute at scale, and the Flip asks the question everyone is circling: Is AGI really here, or is this the best marketing in tech history? Tune in as @PatrickMoorhead & @danielnewmanUV break it all down: 00:02:30 - Reflections on Six Five Pod 300 Episodes 00:03:44 - The Decode | AI Frontier Model Arms Race 00:05:04 - The Decode | Anthropic's Claude Mythos Announcement 00:08:09 - The Decode | OpenAI's SPUD Model Update 00:10:45 - The Decode | Meta's Superintelligence Lab 00:12:11 - The Decode | Intel's Major Developments 00:22:32 - The Decode | Maine's Data Center Construction Ban 00:31:02 - The Flip | AGI Debate: Is It Here? 00:40:04 - Bulls and Bears | Simulated Crossfire on AGI 00:49:03 - Bulls and Bears | Market Trends and Economic Rally Updates 00:52:21 - Bulls and Bears | Supply Chain Concerns and Market Impact

English
1
3
21
5.8K
Ben Rudolph
Ben Rudolph@BenThePCGuy·
@PathToManliness Never too old. 30 on the left (with no kids), 47 on the right (with 10 kids - not a typo). Get your ass in gear.
Ben Rudolph tweet media
English
3
1
11
367
Path To Manliness
Path To Manliness@PathToManliness·
Can’t stand guys who say they’re too old Bro, I’m 38 and I’m about to eat half a pizza and slam a few beers Wake up at 5 AM Run 10 miles Chug a mt dew Bike 20 miles Chug a beer Watch a track meet And I won’t take a nap You’re not old. You’re out of shape and bored
English
60
24
737
31.4K