Global Semiconductor Sales Increase 35.1% from Q1 2026 to Q2 2026

[Daniel Nenni]

Worldwide chip sales in June increase 123.6% year-to-year, 9.7% month-to-month

WASHINGTON, August 6, 2026—The Semiconductor Industry Association (SIA) today announced global semiconductor sales were $403.3 billion during the second quarter of 2026, an increase of 35.1% compared to Q1 of 2026. Global sales were $134.5 billion during the month of June 2026, an increase of 123.6% compared to June 2025 and 9.7% more than sales in May 2026. Monthly sales are compiled by the World Semiconductor Trade Statistics (WSTS) organization and represent a three-month moving average.

“Global chip sales are expected to exceed $1.5 trillion in 2026, with Q2 sales substantially outpacing sales in Q1 2026,” said John Neuffer, SIA president and CEO. “Strong sales across the Americas, the Asia Pacific region, and China continue to drive market growth, underscoring the strength of global chip demand and the vital role semiconductors play in powering the next generation of technology innovation and economic competitiveness.”

Regionally, year-to-year sales in June were up in the Americas (160.9%), Asia Pacific/All Other (124.4%), China (112.8%), Europe (75.2%), and Japan (39%). Month-to-month sales in June increased in China (10.4%), Asia Pacific/All Other (9.8%), the Americas (9.6%), Japan (8.6%), and Europe (7.7%).

For comprehensive monthly semiconductor sales data and detailed WSTS forecasts, consider purchasing the WSTS Subscription Package. For detailed historical information about the global semiconductor industry and market, consider ordering the SIA Databook.

Global Semiconductor Sales Increase 35.1% from Q1 2026 to Q2 2026 - Semiconductor Industry Association

Worldwide chip sales in June increase 123.6% year-to-year, 9.7% month-to-month WASHINGTON, August 6, 2026—The Semiconductor Industry Association (SIA) today announced global semiconductor sales were $403.3 billion during the second quarter of 2026, an increase of 35.1% compared to Q1 of 2026...

www.semiconductors.org

 

[Ole Hoejlund]

In lack of better words: wow…

 

[Daniel Nenni]

In lack of better words: wow…

My guess is that chip prices are increasing more than unit sales. Let's see what Malcolm says.

The Semiconductor Growth Numbers are Insane but the Real World Doesn’t Tally! - Semiwiki

May’s WSTS Report saw March’s total monthly semiconductor sales up 88.1 percent vs. March 2025, albeit down 8.5 percent from February 2026. This month-on-month decline needs to put in the context of February’s record-breaking 25.7 percent monthly growth. This growth, however, was solely...

semiwiki.com

 

[mgp-1]

Spot on Daniel ... and apologies for raining on the SIA parade but it's not only price inflation that's driving this 'growth', its solely attributable to a single, highly specialised, end market application, namely AI hyperscalers. Clearly the numbers are real but the analysis and implication for this growth are superficially false and mis-leading. This is NOT a chip market boom in the broader sense of the meaning ... no other chip market can afford to pay these sky-high ASPs. Just pray the AI boom keeps booming If the music stops, this is going to get nasty ... there is no Plan B!! Be warned, those red and blue curves cannot keep climbing vertical forever.

 

[user nl]

My guess is that chip prices are increasing more than unit sales. Let's see what Malcolm says.

The Semiconductor Growth Numbers are Insane but the Real World Doesn’t Tally! - Semiwiki

May’s WSTS Report saw March’s total monthly semiconductor sales up 88.1 percent vs. March 2025, albeit down 8.5 percent from February 2026. This month-on-month decline needs to put in the context of February’s record-breaking 25.7 percent monthly growth. This growth, however, was solely...

semiwiki.com

Malcom Penn and his colleagues write:

Whilst we absolutely believe AI will eventually transform the world as we currently know it, just as the car, air and rail travel, telecommunications, radio, TV, calculators, computers, the Internet and numerous other breathtaking inventions changed the way we live, work and play, we believe these all take time to materialise and the durable path of the associated technological evolution has yet to emerge.
......


Hard for my gut to accept the premise of Penn (that underlise their prediction for the future) which puts radio and AI on the same level of "world transformation". Some others also seem to observe differences between Radio and AI.

My feeling when reading what others write is that my gut is agreeing more with these latter writers that Penn&Co.

Noone has a cristal ball, let's see in 10 years time whether Penn's gut was right.

 

[Daniel Nenni]

Malcom Penn and his colleagues write:

Whilst we absolutely believe AI will eventually transform the world as we currently know it, just as the car, air and rail travel, telecommunications, radio, TV, calculators, computers, the Internet and numerous other breathtaking inventions changed the way we live, work and play, we believe these all take time to materialise and the durable path of the associated technological evolution has yet to emerge.
......


Hard for my gut to accept the premise of Penn (that underlise their prediction for the future) which puts radio and AI on the same level of "world transformation". Some others also seem to differences between Radio and AI.

My feeling when reading what others write is that my gut is agreeing more with these latter writers that Penn&Co.

Noone has a cristal ball, let's see in 10 years time whether Penn's gut was right.

Malcolm has a record of being right so that is where my money is. I work with EDA/IP start ups and there is definitely and AI bubble in process. One AI EDA company raised $130M after two years of operation. That is unheard of. A dozen or more AI EDA companies will follow. The good news is that EDA is no longer three big dogs eating out of one dog bowl. The bad news is that AI has made it very easy to make an EDA company and get funding so at some point the bubble will pop. Exciting times, no reason to retire now!

 

[user nl]

Malcolm has a record of being right so that is where my money is. I work with EDA/IP start ups and there is definitely and AI bubble in process. One AI EDA company raised $130M after two years of operation. That is unheard of. A dozen or more AI EDA companies will follow. The good news is that EDA is no longer three big dogs eating out of one dog bowl. The bad news is that AI has made it very easy to make an EDA company and get funding so at some point the bubble will pop. Exciting times, no reason to retire now!

DAN, sorry for the misunderstanding and my bad phrasing (should have used AI ). I'm not talking about any short term (1-5 years) bubble popping, that may or may not happen being a little small bubble or somewhat larger bubble popping. That is NOT my point.

There is nothing more difficult than making meanigful predictions in economy and the financial industry. It is NO HARDCORE Science. It is a social acitivity "playing with societies' rent" as some big financial industry banker once framed it. That's all they do "playing with societies' rent", whether you like it or not. Some people like to play with other peoples' surplus.

Penn's premise that RADIO is like AI in 10-20 years time frame was too difficult for my gut to accept...........

(PS the fact that Penn writes that they are contrarian against 99% of the other predictions just underlines my point, it is no MEANINGFUL prediction, it's Penn's gut versus 99% other guts.....if it was scientific stuff that Penn wrote he would not be doing perhaps what he is doing now?)

 

[Daniel Nenni]

DAN, sorry for my misunderstanding and bad phrasing. I'm not talking about any short term bubble popping, that may or may not happen being a little small bubble or somewhat larger bubble popping. That is NOT my point.

There is nothing more difficult than making meanigful predictions in economy and the financial industry. It is NO HARDCORE Science. It is a social acitivity "playing with societies' rent" as some big financial industry banker once framed it. That's all they do "playing with societies' rent", whether you like it or not. Some people like to play with other peoples' surplus.

Penn's premise that RADIO is like AI in 10-20 years time frame was too difficult for my gut to accept...........

Agreed, I'm not a fan of the RADIO analogy either.

From what I see in the trenches AI chips are most certainly a bubble. Nvidia is getting an ASP premium which is causing big customers to make their own. Just about all of the hyperscalers have in-house chip teams now or they are working with the big ASIC companies. There are also dozens of AI chip start-ups in process.

Bottom line: Chip demand will definitely continue to increase but chip ASP has to go down or at least normalize. Since in-house chips are hard to account for the numbers from SIA will be less reliable.

 

[Xebec]

Two observations on the WSTS chart.

1. The latest chart contradicts an older chart of theirs -- chart above shows almost 80% YoY around 2010, while an older chart from WSTS shows just above 60%.
2. The % YoY and revenue numbers don't track at all. An 80% YoY should show more than a .. 20-30% bump in revneue.
3. Assuming there is some accuracy here, the current ramp up looks a lot more severe than even Y2K.

..

Re: Radio and other analogies -- AI "supremacy" has had at least 2 false starts in the last 40-45 years.. The internet even had a "false start" or two before it got really going.. meaning predicting it's success (timing) is .. probably impossible.

 

[Daniel Nenni]

Re: Radio and other analogies -- AI "supremacy" has had at least 2 false starts in the last 40-45 years.. The internet even had a "false start" or two before it got really going.. meaning predicting it's success (timing) is .. probably impossible.

True. During my undergrad I took a LISP class and declared that AI would change the world!!!!! 45 years later I am right!!!!!!!

 

[user nl]

DAN you are even much more ambitious than Penn, you try to predict 45 years ahead! Penn tries 1-5 years ahead only.

It's like the weather and climate, weather prediction is easy, climate prediction is the hard stuff

 

[swka]

Malcolm has a record of being right so that is where my money is. I work with EDA/IP start ups and there is definitely and AI bubble in process. One AI EDA company raised $130M after two years of operation. That is unheard of. A dozen or more AI EDA companies will follow. The good news is that EDA is no longer three big dogs eating out of one dog bowl. The bad news is that AI has made it very easy to make an EDA company and get funding so at some point the bubble will pop. Exciting times, no reason to retire now!

bubble in the sense of many of them will fail is true. happens everytime a disruption happens. anyone remember the dozens of DFM startup 20 year ago. This time, like everything else AI touches, we have outsized funding round, some are 100x more than what "traditional" eda startup ever gets.

as for retirement Dan, i have learnt this is really a mindset. as long as you control your time 100%, it is retirement. For high drive people, retirement is never just travelling, golfing, etc ...