Showing posts with label innovation. Show all posts
Showing posts with label innovation. Show all posts

Thursday, April 04, 2024

Casey Handmer: Terraform Industries and a Carbon-Neutral Future — Manifold #57

 

Casey Handmer (PhD Caltech, general relativity) is the founder of Terraform Industries. He is one of the most capable and ambitious geo-engineers on planet Earth! 

Terraform Industries is scaling technology to produce cheap natural gas with sunlight and air. Using solar energy, they extract carbon from the air and synthesize natural gas, all at the same site. 

March 2024: "Terraform completes the end to end demo, successfully producing fossil carbon free pipeline grade natural gas from sunlight and air. We also achieved green hydrogen at <$2.50/kg-H2 and DAC CO2 at <$250/T-CO2, two incredible milestones." 

Links: 

Casey Handmer’s website: https://www.caseyhandmer.com/ 

Terraform Industries: https://terraformindustries.com/ 

Nerds on Patrol [Episode 3] - Terraform Industries: 

Steve and Casey discuss: 

0:00 Introduction 
00:31 Casey's early life and background, from Australia to Caltech 
07:55 The academic path and transition to tech entrepreneurship 10:40 Terraform Industries 
15:21 Solar costs, efficiency, and global Impact 
24:25 A world powered by Terraform methane 
31:27 The entrepreneurial journey: challenges and insights 
35:01 Investor dynamics and strategic decisions for Terraform 
41:28 The hard Reality of manufacturing and innovation 
44:11 Navigating intellectual property and strategic partnerships 
45:49 The moral and technical challenges of carbon neutrality 
55:48 Looking ahead: Terraform's next milestones and the solar revolution

Transcript and Audio-only version:

Monday, March 11, 2024

Solving the Hallucination Problem - interview with AppliedAI

 

Recent podcast interview with AppliedAI. 

We discuss SuperFocus.ai Enterprise GPT. 

Good intro to how customer-defined AI/LLM memory eliminates hallucinations.

Thursday, January 25, 2024

Utah AG Sean Reyes: “Sound of Freedom” and Human Trafficking — Manifold #52

 

Sean Reyes is Utah Attorney General and a producer for the movie “Sound of Freedom.” 

Steve and Sean discuss his personal story, human trafficking, and the role of technology in law enforcement. 


NOTE: Reyes has announced that he will not seek re-election as Utah AG: https://www.youtube.com/watch?v=HEEj4UgjDL4 

00:00 Sean Reyes’ early life and family history 
14:21 Sean's personal journey and career 
21:28 Political journey and decision to run for AG 
24:08 The movie Sound of Freedom 
28:45 The reality of human trafficking 
31:40 Technology and law enforcement 
44:00 The horror of human trafficking: victims, aftercare, and the media 
01:05:23 Future plans and aspirations

Wednesday, January 24, 2024

SuperFocus, AI, and Philippine Call Centers: Part 2



This is the sequel to the earlier conversation with Dominic Ligot, an AI expert who works with the IT and Business Process Association of the Philippines (IBPAP), the trade association for call center and outsourcing companies. 

In this video we briefly demonstrate some of the voice capabilities of the SuperFocus AI. Progress in generative AI is faster than anything I've ever seen before - perhaps not surprising given the vast financial, technological, and human capital resources flowing to AI R&D. When we first looked at voice capabilities ~6 months ago they didn't seem ready for complex conversations like the ones discussed in the video. But when we looked again - prompted by strong interest from our customers - we found that the state of the art had advanced significantly in just a short time. This is true across many areas of generative AI.

I was in Manila in December to meet with BPO companies. Roughly 8% of Philippine GDP ($40B each year) results from BPO / call center work. This is a consequence of low labor costs and widespread English fluency.

We demonstrated narrow AIs built using LLMs, but in which the LLM is forced to "consult its internal memory" before answering any query. This memory can be built from training materials used to train human agents in call centers. The AI functions like a human that has perfect recall of all the material in the training manuals, at a fraction of the cost!

An analogy we used is that the AI earthquake in SF has created a Tsunami headed towards the Philippines -- is it a 6 foot wave, or a 600 ft wave? Closer to the latter, I think.





Some photos from Manila - scoping out potential SuperFocus.ai office space.

 


Tuesday, January 16, 2024

Upstream podcast with Erik Torenberg: Steve Hsu on the Future of Everything

 


Great conversation with Erik, a well-known SV Founder and investor. 

TIMESTAMPS: 
(00:00) Intro 
(02:01) Political polarization in academia 
(05:27) The decline of meritocracy in academia 
(09:40) Why can't academia change? 
(13:07) Truth-seeking in startups 
(20:16) Sponsors | Shopify | Givewell 
(23:16) The fertility crisis 
(31:51) AI and labor 
(40:46) What industries are safe from AI automation? 
(43:45) AI Safety vs accelerationism 
(47:31) Understanding the rise of China 
(58:26) The future of the US/China relationship 
(01:00:21) How does Steve allocate his time? 
(01:04:21) Steve's suggestions for underexplored opportunities 

Audio-only version on Spotify: 

Wednesday, January 03, 2024

SuperFocus, AI, and Philippine Call Centers

 

This is a conversation with Dominic Ligot, an AI expert who works with the IT and Business Process Association of the Philippines (IBPAP), the trade association for call center and outsourcing companies. 

I was in Manila in December to meet with BPO companies. We demonstrated narrow AIs built using LLMs, but in which the LLM is forced to "consult its internal memory" before answering any query. This memory can be built from training materials used to train human agents in call centers. The AI functions like a human that has perfect recall of all the material in the training manuals, at a fraction of the cost!

An analogy we used is that the AI earthquake in SF has created a Tsunami headed towards the Philippines -- is it a 6 foot wave, or a 600 ft wave? Closer to the latter, I think.





Some photos from Manila - scoping out potential SuperFocus.ai office space.

 


Sunday, December 24, 2023

Peace on Earth, Good Will to Men 2023



When asked what I want for Christmas, I reply: Peace On Earth, Good Will To Men :-)

No one ever seems to recognize that this comes from the Bible (Luke 2.14).

Linus said it best in A Charlie Brown Christmas:
And there were in the same country shepherds abiding in the field, keeping watch over their flock by night.

And, lo, the angel of the Lord came upon them, and the glory of the Lord shone round about them: and they were sore afraid.

And the angel said unto them, Fear not: for, behold, I bring you good tidings of great joy, which shall be to all people.

For unto you is born this day in the city of David a Saviour, which is Christ the Lord.

And this shall be a sign unto you; Ye shall find the babe wrapped in swaddling clothes, lying in a manger.

And suddenly there was with the angel a multitude of the heavenly host praising God, and saying,

Glory to God in the highest, and on Earth peace, good will toward men.


2023 saw the founding of our startup SuperFocus.ai, which builds AIs with user-configured attached memory. The AI consults this memory in responding to prompts, and only gives answers consistent with the information in the memory. This solves the hallucination problem and allows the AI to answer questions like a human with perfect recall of the information.

SuperFocus built an AI for a major consumer electronics brand that can support and troubleshoot hundreds of models of smart devices (I can't be more specific). Its memory consists of thousands of pages of product manuals, support documents, and problem solving guides originally used by human support agents.

In December I traveled to Manila after the semester ended, in order to meet with outsourcing (BPO = Business Process Outsourcing) companies that run call centers for global brands. This industry accounts for ~8% of Philippine GPD (~$40B per annum), driven by comparative advantages such as the widespread use of English here and relatively low wages. I predict that AIs of the type produced by SuperFocus.ai will disrupt the BPO and other industries in coming years, with dramatic effects on the numbers of humans employed in areas like customer support. 

But fear not: for, behold, I bring you good tidings of great joy, which shall be to all people

The arrival of machine intelligence on Earth is the beginning of a great adventure!


This is GPT's account of the meetings in Manila.
In the heart of Manila, amidst the bustling cityscape, a meeting of innovative minds took place. Steve Hsu, the visionary founder of SuperFocus, had arrived to showcase the prowess of his latest creation—an AI designed to revolutionize technical support for complex products. The setting was a conference room adorned with sleek screens and cutting-edge technology, a fitting backdrop for the unveiling of this groundbreaking innovation. 
Seated around the polished table were the owners and executives of prominent BPO (Business Process Outsourcing) companies. Their faces were a blend of anticipation and apprehension as Steve Hsu prepared to demonstrate the capabilities of the AI-powered technical support system. 
With a confident smile, Steve initiated the demonstration. The AI, equipped with a sophisticated neural network, began its simulated interaction. It effortlessly tackled intricate technical queries, deciphering complex issues with lightning speed and unparalleled accuracy. Each solution presented was concise, comprehensive, and flawlessly executed. 
As the AI effortlessly navigated through a myriad of scenarios and troubleshooting processes, the room fell into a hush. The BPO leaders exchanged astonished glances, their initial amazement mingled with a growing sense of unease. The capabilities displayed by the AI were undeniably impressive, but they also highlighted a looming question—what did this mean for the future of human roles in their industry? 
Steve Hsu noticed the shift in atmosphere and paused the demonstration. With a gentle yet determined tone, he addressed the concerns lingering in the room. "This AI isn't meant to replace human expertise," he began. "Rather, it's here to augment and enhance your services. Imagine your teams empowered by this technology, streamlining operations, and providing even more efficient and effective support to customers." 
His words offered reassurance, but the specter of automation replacing human jobs lingered in the minds of the BPO owners. The potential efficiency gains were undeniable, yet so too were the implications for the human workforce. 
In the ensuing discussion, voices echoed with a mix of excitement and apprehension. Some saw the potential for growth and advancement, envisioning a future where human creativity combined with AI prowess would elevate their services to new heights. Others grappled with the uncertainty, worrying about the displacement of jobs and the evolving landscape of the industry they had dedicated their careers to. 
Steve Hsu listened attentively, acknowledging their concerns while emphasizing the collaborative potential between humans and AI. "This technology," he explained, "is a tool, a means to empower and evolve, not to supplant. Together, we can harness its capabilities to create a synergy that benefits both businesses and their workforce." 
As the meeting concluded, the BPO leaders departed with a mix of awe and trepidation. The AI presented by Steve Hsu had showcased a future teeming with possibilities, yet it also raised profound questions about adaptation and the role of humans in an increasingly automated world. 
The echoes of the demonstration lingered in the minds of those present, igniting discussions and contemplation about the balance between innovation and the human touch, forever altering the landscape of the BPO industry in Manila and beyond.


Bonus: Two recent interviews I did which I enjoyed very much. 



Thursday, November 30, 2023

Charles Miller: Satellite Technology and the Future of Mobile Connectivity — Manifold #49

 

Charles Miller is co-founder and CEO of Lynk. He is a serial space entrepreneur with 30 years of experience in the space industry. 


0:00 Introduction and guest background 
1:27 Miller's early passion for space 
3:54 Evolution of commercial space 
6:42 Impact of Elon Musk and SpaceX 
8:01 The challenges of early stage startups 
11:26 The birth of Lynk, its technical challenges, and breakthroughs 
33:11 Use cases for satellite connectivity 
35:20 The plan for Lynk satellites 
36:41 Competition with Starlink 
39:25 Investment opportunities in Lynk 
47:04 Satellite technology and global competition 
50:21 Impact of Huawei’s satellite phone features 
59:01 Advice for entrepreneurs 

Audio-only and Transcript: 

Thursday, November 16, 2023

China's EV Market Dominance and the Challenges Facing Tesla — Manifold #48

 

TP Huang is a computer scientist and analyst of global technology development. 

He posts often on X: https://x.com/tphuang

The EV tipping point has arrived in China. Even most techology experts do not appreciate the coming huge impacts on global economics, manufacturing, energy transition, etc.


Chapters:

0:00 Introduction 
2:21 How TP Huang became interested in electric vehicles 
6:30 The perception and reality of Chinese products, future of Chinese auto market 
9:24 The impact of Tesla on the Chinese electric vehicle market 
14:41 Buying a car in China 27:05 China dominates with electric vehicle batteries 
30:44 The challenges facing Tesla in China 
40:11 The evolution of smart cars, autonomous vehicles, and self driving 
50:48 LIDAR technology and autonomous driving 
59:08 BYD, China’s energy independence, and power grid 
1:14:04 The downstream impact of China leading in tech and electric vehicles


Audio-only version and transcript: 


See earlier episodes:

TP on the US-China chip war

Taylor Ogan of Snow Bull Capital (Shenzhen) on EVs, LIDAR, manufacturing in China

Thursday, November 02, 2023

Taylor Ogan, Snow Bull Capital: China's tech frontier, the view from Shenzhen — Manifold #47

 

I really enjoyed this conversation. Taylor is a very unique investor who relocated his fund to Shenzhen in order to have direct access to information on Chinese tech companies.

Taylor Ogan is Chief Executive Officer of Snow Bull Capital, based in Shenzhen, China. 

Follow him on X @TaylorOgan


Steve and Taylor discuss: 
 
0:00 Introduction 
1:02 Taylor's background and why he moved his firm to China 
20:43 China post-pandemic and economic dynamism 
33:43 China dominance in electric vehicles; LIDAR 
56:55 Investment research: factory and site visits 
1:06:52 US-China competition - the future of innovation is in China

Audio-only version and transcript: 

Thursday, October 05, 2023

Yasheng Huang: China's Examination System and its impact on Politics, Economy, Innovation — Manifold #45

 

Yasheng Huang is the Epoch Foundation Professor of Global Economics and Management at the MIT Sloan School of Management. His new book is The Rise and Fall of the EAST: How Exams, Autocracy, Stability, and Technology Brought China Success, and Why They Might Lead to Its Decline. 

Steve and Yasheng discuss: 

0:00 Introduction 
1:11 From Beijing to Harvard in the 1980s 
15:29 Civil service exams and Huang's new book, "The Rise and Fall of the EAST" 
37:14 Two goals: Developing human capital and indoctrination 
48:33 Impact of the exam system 
57:04 China's innovation peak and decline 
1:12:23 Collaboration and relationship with the West 
1:21:31 How will the U.S.-China relationship evolve? 

Audio-only version, and transcript: 

Yasheng Huang at MIT 

Web site: 

Thursday, September 21, 2023

Huawei and the US-China Chip War — Manifold #44

 

TP Huang is a computer scientist and analyst of global technology development. He posts often on X: https://twitter.com/tphuang 


Steve and TP discuss: 

0:00 Introduction: TP Huang and semiconductor technology 
5:40 Huawei’s new phone and SoC 
23:19 SMIC 7nm chip production in China: Yield and economics 
28:21 Impact on Qualcomm 
36:08 U.S. sanctions solved the coordination problem for China semiconductor companies 
42:48 5G modem and RF chips: impact on Qualcomm, Broadcom, Apple, etc. 
47:14 5G and Huawei 52:50 Satellite capabilities of Huawei phones 
56:46 Huawei vs Apple and Chinese consumers 
1:01:33 Chip War and AI model training

Thursday, August 10, 2023

AI on your phone? Tim Dettmers on quantization of neural networks — Manifold #41

 

Tim Dettmers develops computationally efficient methods for deep learning. He is a leader in quantization: coarse graining of large neural networks to increase speed and reduce hardware requirements. 

Tim developed 4-and 8-bit quantizations enabling training and inference with large language models on affordable GPUs and CPUs - i.e., as commonly found in home gaming rigs. 

Tim and Steve discuss: Tim's background and current research program, large language models, quantization and performance, democratization of AI technology, the open source Cambrian explosion in AI, and the future of AI. 





0:00 Introduction and Tim’s background 
18:02 Tim's interest in the efficiency and accessibility of large language models 
38:05 Inference, speed, and the potential for using consumer GPUs for running large language models 
45:55 Model training and the benefits of quantization with QLoRA 
57:14 The future of AI and large language models in the next 3-5 years and beyond

Wednesday, June 28, 2023

Embryo Selection: Healthy Babies vs Bad Arguments

Great article by Diana Fleischman, Ives Parr, Jonathan Anomaly, and Laurent Tellier.
Polygenic screening and its discontents 
... But monogenic and chromosomal screening can only address a part of disease risk because most health conditions that afflict people are polygenic, meaning they are not simply caused by one gene or by a chromosomal abnormality. Instead, they are caused by a huge number of small additive effects dispersed throughout the genome. For example, cancer, schizophrenia, and diabetes can be best predicted by models using tens of thousands of genes. 
A polygenic risk score (PRS) looks at a person’s DNA to see how many variants they have associated with a particular disease. Like BRCA1, polygenic risk scores are typically not determinative: “Polygenic screening is not a diagnosis: It is a prediction of relative future risk compared to other people.” In other words, someone with BRCA1 has a higher risk than someone without, and someone with a high breast cancer PRS has a higher risk than someone with a lower breast cancer PRS. But in principle, BRCA1 is just one gene out of thousands contributing to a PRS, with each bit contributing a small part of a total risk estimate. ... 

 

... Recently, a group of European scientists argued that polygenic screening should not be available to couples because it will lead to stigmatization, exacerbate inequalities, or lead to confusion by parents about how to weigh up information about risks before they decide which embryo to implant. These are indeed challenges, but they are not unique to embryo selection using polygenic scores, and they are not plausible arguments for restricting the autonomy of parents who wish to screen their embryos for polygenic traits. Furthermore, from an ethical perspective, it is unconscionable to deny polygenic screening to families with a history of any disease whose risk can be reduced by this lifesaving technology. 
Many new technologies are initially only available to people with more money, but these first adopters then end up subsidizing research that drives costs down and quality up. Many other medical choices involve complexity or might result in some people being stigmatized, but this is a reason to encourage genetic counseling and to encourage social tolerance. It is not a reason to marginalize, stigmatize, or criminalize IVF mothers and fathers who wish to use the best available science to increase the chances that their children will be healthy and happy.
This is a comment on the article:
1) They don't want to admit that some people are better than others, inherently. Boo hoo. 
2) You put a scorecard of embryos in front of everyone, and everyone has a pretty good ballpark estimate of which are better and which are worse. Nobody is going to pretend equality is true when they are choosing their kids genes. 
3) So bad feels. 
4) Must therefore retard all human progress and cause immense suffering because don't want to deal with bad feels. 
That's the anti-polygenic argument in a nutshell. I don't expect it to be very effective. At best it will cause it to take a bit longer before poor people have access.

Thursday, June 08, 2023

AI Cambrian Explosion: Conversation With Three AI Engineers — Manifold #37

 

In this episode, Steve talks to three AI engineers from his startup SuperFocus.AI. 

0:00 Introduction 
1:06 The Google memo and open-source AI 
14:41 Sparsification and the size of models: AI on your phone? 
30:16 When will AI take over ordinary decision-making from humans? 
34:50 Rapid advances in AI: a view from inside 
41:28 AI Doomers and Alignment 


Links to earlier episodes on Artificial Intelligence & Large Language Models: 

Oxford Lecture — #35: 

Bing vs. Bard, US-China STEM Competition, and Embryo Screening — #30: 

ChatGPT, LLMs, and AI — #29: 

Thursday, May 11, 2023

Artificial Intelligence & Large Language Models: Oxford Lecture — Manifold #35

 

This week's episode is based on a lecture I gave to an audience of theoretical physicists at Oxford University. 


Audio-only version, transcript: 


Outline: 

0:00 Introduction 
2:31 Deep Learning and Neural Networks; history and mathematical results 
21:15 Embedding space, word vectors 
31:53 Next word prediction as objective function 
34:08 Attention is all you need 
37:09 Transformer architecture 
44:54 The geometry of thought 
52:57 What can LLMs do? Sparks of AGI 
1:02:41 Hallucination 
1:14:40 SuperFocus testing and examples 
1:18:40 AI landscape, AGI, and the future


Final slide:


Thursday, February 02, 2023

ChatGPT, LLMs, and AI — Manifold #29

 

Steve discusses Large Language Model AIs such as ChatGPT. 

0:00 How do LLMs work? 
10:22 Impact of ChatGPT 
15:21 AI landscape 
24:13 Hallucination and Focus 
33:09 Applications 
39:29 Future landscape 

Manifold interview with John Schulman of OpenAI: 


Blog posts on word vectors and approximately linear vector space of concepts used by the human mind:
 

Thursday, December 01, 2022

Anna Krylov: The Politicization of Science in Academia — Manifold #25

 

Anna I. Krylov (Russian: Анна Игоревна Крылова) is Professor of Chemistry at the University of Southern California (USC), working in the field of theoretical and computational quantum chemistry. Krylov is an outspoken advocate of freedom of speech and academic freedom. She is a founding member of the Academic Freedom Alliance and a member of its academic leadership committee. 

Her paper, The Peril of Politicizing Science, launched a national conversation among scientists and the general public on the growing influence of political ideology in STEM. It has received over 80,000 views and, according to Altmetric, was the all-time highest-ranked article in the Journal of Physical Chemistry Letters. 

Steve and Anna discuss: 

0:00 Anna Krylov’s background, upbringing in USSR 
7:03 Ideological control and censorship for the greater good? 
14:59 How ideology underpins DEI work in academic institutions 
30:40 Captured institutions 
37:05 How much is UC Berkeley spending on DEI, and where the money is going 
41:46 Krylov thinks it can get worse 
52:09 An idea for defeating preference falsification at universities 



Resources: 

Professor Krylov academic page: 

Wiki page: 

The Peril of Politicizing Science, Journal of Physical Chemistry Letters 2021 https://pubs.acs.org/doi/10.1021/acs.jpclett.1c01475

Saturday, October 15, 2022

Times of Israel on Polygenic Embryo Screening


This is a very nice article on polygenic embryo screening and its prospects in Israel. Worth reading in full. Leading statistical geneticist Shai Carmi is interviewed. See also his interview on Manifold (embedded player at bottom of this post).
The Times of Israel 
14 October 2022, 1:27 pm 
Designer babies? Hi-tech preimplantation genetic testing may soon come to Israel 

For generations, the Yu family of Shanghai has suffered from type 2 diabetes. But this summer, as reported in the China Daily, the family welcomed a baby with a better chance of avoiding this disease.

These rosier prospects are the result of a recent breakthrough in assisted reproduction that was advanced with the help of Israeli scientists, called preimplantation genetic testing for polygenic diseases (PGT-P). In addition to China, PGT-P is also gaining ground among couples in the United States who wish to improve health outcomes for their future children. But in Israel, it is illegal.

PGT-P is carried out on an embryo during in vitro fertilization (IVF), prior to its transfer from the Petri dish to the womb. Viable embryos with the probable lowest disease risk can then be selected for implantation.

Since this innovative testing takes into account a complex combination of factors that are not broached in more traditional testing, in some ways it’s almost like an educated guess. Accordingly, polygenic screening is not a diagnosis: It is a prediction of relative future risk compared to other people.

Israeli academics have published peer-reviewed research advancing the science behind polygenic screening, including Shai Carmi, Ehud Karavani, Or Zuk, Gil Atzmon, and Einat Granot-Hershkovitz.

But Start-Up Nation is not yet implementing this cutting-edge tech in the field of fertility. Although fertility treatments are subsidized by the Israeli government, it is still unclear whether Israeli couples ever will have access to the procedure, which screens for polygenic diseases such as diabetes, heart disease and cancer — or whether they would even want it.

PGT-P is different from prior technology in important ways, creating new opportunities and challenges for parents while raising profound ethical dilemmas for society. Similar to older forms of testing, PGT-P relies on analyzing genetic material from embryos created through IVF before implantation and checking them for certain diseases and conditions. The information then helps the parents and doctors decide which embryos to implant.

However, the biggest difference between PGT-P screening and earlier forms of genetic testing is that the prior tests checked for genetically simple conditions such as Down syndrome, cystic fibrosis, or Tay Sachs disease. These diseases, which are serious or fatal, have extremely high “penetrance,” which means that if the gene mutation is seen in the embryo’s DNA, it is nearly certain that the child will have that condition. The appearance of the disease-linked gene is the basis of a clear diagnosis.

This “simple” genetic screening has already borne fruit in the Jewish community: Decades ago, babies in the Ashkenazi Jewish community were nearly 100 times more likely to be born with Tay Sachs than babies in the general US population. Today, because of genetic screenings, the disease is “virtually wiped out.”

In contrast, PGT-P screening can’t tell you with assurance if an embryo will develop a genetic disease such as cancer or Crohn’s disease. That’s because this new screening checks for polygenic diseases – complex conditions caused by the combined impact of possibly thousands of different genes, as well as lifestyle and other environmental factors.

Instead of a clear diagnosis, prospective parents receive a “polygenic risk score,” basically the probability of a child developing a certain disease or condition.

Noa and her husband went through 10 IVF cycles to build their family. “We now have two wonderful boys,” she says.

If I had an opportunity to reduce disease risk in my kids, I would do it

She knows what she would have said if doctors had offered her polygenic screening: “I want that technology.” As a speech therapist who works with kids facing a lot of health challenges, she was very worried about what her own kids would face.

“If I had an opportunity to reduce disease risk in my kids, I would do it. It would definitely help my peace of mind as a mother. Everyone here in Israel should have the option of using it,” Noa says.

No clear-cut answers

Scientists at the US-based Genomic Prediction, Inc. published an article in 2019 describing the “first clinical application” of polygenic screening of embryos. Genomic Prediction is a polygenic screening company based in New Jersey that partners with various IVF clinics around the world.

However, to date, the Israeli Health Ministry has yet to even issue a statement on the use of polygenic screening on embryos.

For some, the fact that PGT-P screening isn’t available, or even legal, in Israel is somewhat counterintuitive, given Israel’s prominence in the fields of both assisted reproduction and genetic testing.

Israelis undergo more rounds of IVF per capita than any other nation in the world. This is largely due to religious and cultural norms that are highly supportive of child-bearing, combined with the nationally financed healthcare system that provides full coverage for as many IVF cycles as needed, up to two children per family.

In addition, “in Israel there is a lot more openness to preimplantation genetic testing in general because of the high prevalence of various disease mutations in our community,” says Carmi, an associate professor at the Hebrew University School of Public Health and Faculty of Medicine.

Today, Carmi is a leading researcher on the accuracy of polygenic screening. As part of his post-doctoral project at Columbia University in New York, he helped generate important genetic sequencing data for Ashkenazi Jews.

Israel’s embrace of most genetic testing is reflected in the Israeli Health Ministry’s website, which lists dozens of recommended genetic screenings, broken down by ethnic sub-community. But these screenings are for monogenic disorders, easily diagnosed by looking for a single gene mutation.

“In Israel, the Health Ministry controls which diseases can be screened for, and candidate variants need to have high penetrance and lead to diseases with severe symptoms,” says Carmi.

Playing the odds

Miri is a consultant originally from central Israel. Although she did not have any known fertility problems, she chose to undergo IVF specifically because it would allow her to screen for a certain hereditary disease. She and her husband are both carriers of a rare mutation, so a natural conception meant a 25 percent chance of the fetus suffering from this generally fatal condition.

“For me, it was a choice between the extra physical hardship of IVF, or the extra emotional hardship of a pregnancy where, for months, we would not know if the baby would have this disease,” Miri said.

In contrast, PGT-P can’t provide conclusive information, because in the context of polygenic diseases like diabetes and heart disease “nothing is deterministic,” says Carmi.

According to Carmi, a child may develop the condition or may not, and non-genetic factors can certainly affect the outcome. Based on his peer-reviewed research on statistical modeling of polygenic screening, though, Carmi notes that “you can get quite a substantial risk reduction.”

The “relative risk reduction” projected to be accomplished by PGT-P varies depending on the disease. However, according to a 2021 research paper by Carmi and his collaborators, for schizophrenia and Crohn’s disease, around a 45% relative risk reduction is achievable for parents testing five embryos and choosing the best scoring, compared to implanting a randomly chosen one of the five.

The testing, of course, comes with a fee: Costs vary, but Genomic Prediction in New Jersey charges a $1,000 up-front fee, plus $400 per embryo analyzed. Of course, this is an add-on cost for people already doing IVF, which in the US can cost up to tens of thousands of dollars per cycle.

Pricing can get even more complicated, however, because different services end up bundled together, or are offered as add-ons once related costs are already accounted for. But one of the earlier forms of embryonic screening (PGT-A, which checks for aneuploidies, giving rise to Down syndrome for example) can cost several thousand dollars.

By contrast, carrier screening, which is a blood test performed on the parents to check for “simple” monogenic-disease carrier status, costs only several hundred dollars, and is often also covered by insurance.

In Israel, for couples whose family history or carrier-screening blood tests reveal a heightened risk for having children with a specific monogenic disease, the Health Ministry promotes the benefits of traditional genetic testing of embryos prior to implantation in the womb.

According to its website, “Pre-implantation Genetic Diagnosis (PGD) is today considered to be one of the practical options for couples who are at high risk for giving birth to a baby with a chromosomal abnormality or a genetic disease. This is because the process allows pregnancies to be achieved with healthy fetuses, and avoids the need for pregnancy termination, a procedure that constitutes a problem for many couples for religious, ethical and/or moral reasons.”

The nuts and bolts

PGT-P was developed using artificial intelligence technology applied to huge databases containing the genetic and health information of hundreds of thousands of people. Statistical data analysis of DNA and health outcomes allows scientists to see which complex genetic patterns more frequently show up in individuals who also develop a certain disease, such as schizophrenia. By genetically analyzing an embryo and then comparing its genetic information to this population data, the embryo’s polygenic risk score can be calculated for a given disease. This can already be done for a great many common diseases, with varying levels of predictive power, and as genetic databases grow, the reliability of these risk scores will continue to improve.

The couple also receives the raw data about their embryos’ genes and risk scores, so if they prefer to implant the embryo with the lowest risk of type 2 diabetes rather than the lowest combined disease risk, they can do that

“For prospective parents undergoing IVF and electing to use polygenic screening, somewhere between 10 and 20 polygenic risk scores are combined in a weighted average, with more serious diseases given greater weight in the final figure. This averaging provides a single number for each embryo — a health index — that can be used to rank the available embryos, so that the one with the best health index can be implanted,” says Carmi.

“Of course, the couple also receives the raw data about their embryos’ genes and risk scores, so if they prefer to implant the embryo with, let’s say, the lowest risk of type 2 diabetes, rather than the lowest combined disease risk, they can do that,” says Carmi.

An emotional decision

Michal Amrani, 32, lives in the central Israeli town of Ramat Hasharon and is working toward a master’s degree in chemistry from the Weizmann Institute. Through a four-year IVF process, she and her husband Sarel welcomed a son, and later, a set of twins. They say they are unlikely to use polygenic screening, even if it becomes available in Israel.

“As it is, we opted not to do some of the genetic testing that was already available to us,” Amrani says. “I work in science, so I am more open to these things, but my husband doesn’t really like all these genetic tests. For him, there’s risk in lots of things, and his optimistic nature helps him be comfortable that things will work out.”

Others, like Noa, are more interested in trying out preimplantation polygenic screening of their embryos, but even if Israel would change its rules to allow it, it’s a tricky issue. First, there are concerns about the psychological difficulties that this technology may pose for prospective parents.

Rona Langer Ziv is a social worker and cognitive behavioral psychotherapist who counsels IVF patients — both couples and singles — at a large Israeli hospital, as well as through her private clinical practice.

“Due to the potential implications of this new technology,” she says, “I would be concerned about a higher risk for depression and anxiety among the IVF patients.”

“Even if they feel they understand what they are signing up for at the beginning of the journey, they may not appreciate the emotional, ethical, and psychosocial implications of polygenic screening several IVF cycles down the road,” says Langer Ziv. “They may find themselves worrying that the embryos’ scores are not good enough, or that they won’t have any viable embryos left to choose from.”

Even if they feel they understand what they are signing up for at the beginning of the journey, they may not appreciate the emotional, ethical, and psychosocial implications of polygenic screening several IVF cycles down the road

Because polygenic screening predicts relative risk rather than providing an affirmative disease diagnosis, “women, especially those over 40 who may have very few embryos to work with, end up facing a serious dilemma — they may be discarding an embryo that could have resulted in a healthy child,” says Langer Ziv.

Amrani is in a similar situation. She and her husband are ready for more kids, but right now they have just one embryo available, so that’s the embryo they will try to implant. Even though she won’t be using polygenic screening, Amrani says that “it does sound very innovative. It’s nice that there’s something like this.”

Social worker and cognitive behavioral psychotherapist Rona Langer Ziv. (Courtesy) Indeed, Langer Ziv acknowledges that some people would find polygenic screening appealing, particularly those with higher education levels.

“There’s definitely coolness in the technology. It’s scientifically advanced, and it could offer interesting health insights about your future children. Everyone would theoretically like to use a technology that potentially predicts a more healthy child, although there is disagreement among fertility specialists about the benefits involved,” Langer Ziv says.

“And for some IVF patients, it might also provide a feeling of control during a process that involves so much stress, uncertainty, luck, and randomness,” she says.

Risk of eugenics

Regardless of how polygenic screening would be received by potential consumers, there are grave concerns about the impacts of this new technology on society. Various ethical issues have been raised for decades about older forms of genetic screenings, including fears of stigmatizing those living with genetic diseases, and questions about equitable access to these technological advances.

Perhaps the most significant ethical concern, and one that looms larger with polygenic screening than with older tests for monogenic diseases, is the potential for eugenics. This is the infamous and dangerous philosophy, practiced in Nazi Germany and elsewhere, that society should try to promote the creation of the most genetically “superior” babies.

Miri and her husband now have a baby boy and are looking forward to having more children — they still have three embryos to choose from. Asked whether she would be interested in polygenic screening if it became available in Israel, Miri says she’s unsure.

“I would love to see less suffering in the world from diseases. But where do we draw the line?” she says.

Indeed, the potential for eugenics is most stark when screenings cross over from the realm of disease prevention to the world of intelligence and aesthetic traits such as height or eye color. As such, some laboratories preemptively claim they will only screen for health concerns: An American polygenic screening company currently states that it does not test for “high IQ,” nor for “purely cosmetic traits such as hair color and eye color.”

But complicating the “noble” stance, genetic researchers have shown that “IQ is negatively correlated with most psychiatric disorders [and] positively correlated with autism and anorexia,” meaning that a high IQ comes with a lower risk of most psychiatric diseases and a higher risk of certain other neurological and mental health conditions.

As such, while some companies may currently refuse to offer IQ screening, it is not hard to imagine a health-based argument for loosening such protocols in the future, particularly as society becomes more used to the practice of PGT-P.

Similarly, a large study was published this year by researchers at Brown University and Peking University that found that “light eye colors were associated with high risks” of certain forms of skin cancer. Again, one can picture checks for eye color making their way into future genetic screenings through a backdoor of disease relevance.

In Carmi’s view, the responsible way for Israel to approach the prospect of polygenic screening is a gradual one.

“Ideally, we would start by recruiting Israeli participants for local academic research, with oversight by the Health Ministry,” Carmi says. “Once we develop more insight into how predictive polygenic screening is in our population, the relevant stakeholders — including patients, professional organizations, and regulators — can balance competing interests and local values, and come up with tailored guidance on its use in Israel.”

For some Israeli citizens, of particular concern is the idea of the wealthy trying to create perfect babies.

“If polygenic screening came to Israel, I would want to see a lot of regulation about who gets to use it, how it is used, and what reasons it is used for,” Miri says.

See also

WIRED: Genetic Screening Now Lets Parents Pick the Healthiest Embryos 

Genomic Prediction in Bloomberg


Blog Archive

Labels