Archive for the ‘Alphago’ Category

One step closer to the Matrix: AI defeats human champion in Street … – TechRadar

Researchers from the Singapore University of Technology and Design (SUTD) created a new software centered around reinforcement learning and phase-change memory thats designed to understand complicated movement design.

Previous work has applied this kind of deep learning to other games like Chess or Go, but they decided instead to expose the D-PPO algorithm to the rigors of Street Fighter Champion Edition II. The SUTD researchers trained its SF-R2 AI player on two days of consecutive play against the computer, before letting it loose on a human participant who the AI-powered system beat comfortably.

The work has implications for movement science more broadly, according to the research paper, and can possibly be fed into improving robotics and autonomous vehicles, for example. It paves the way for broadly applicable training in fields where machines may observe human norms and attempt to replicate and outperform them.

One of the major milestones that AI researchers have used to measure the effectiveness of the systems theyve built is by letting them compete with human players in different kinds of games. This has been happening for some time.

In 2017, an Alpha Go AI built by DeepMind beat the number-one human Go player in the world for the second time, following the first victory over Fan Hui the previous year. Microsofts AI, in June, achieved the worlds first perfect Ms. Pac-Man score, and in August we saw an OpenAI engine beating the best Dota 2 players of the time.

This latest milestone besting a Street Fighter champion was made possible due to reinforcement learning as well as phase-change memory. First developed by HP, this is a form of nonvolatile memory achieved by using electrical charges to change areas on chalcogenide glass. Its much faster than commonly used Flash memory.

"Our approach is unique because we use reinforcement learning to solve the problem of creating movements that outperform those of top human players, said principal investigator Desmond Loke to TechXplore. This was simply not possible using prior approaches, and it has the potential to transform the types of moves we can create.

Continue reading here:
One step closer to the Matrix: AI defeats human champion in Street ... - TechRadar

The Vanishing Frontier – The American Conservative

Today, technology, state power, and development intertwine in a single Gordian knot that no liberal can untangle. After a few decades of geopolitical quiet, we have come to realize that threats to sovereignty have become more insidious, while decisions that lead to its preservation require subtle judgment, especially where the matters concern technology. Awareness of this new state of affairs is particularly acute in the two great powers: the U.S. and China.

The uniqueness of the circumstances was recognized by both Donald Trump and Xi Jinping. In Innovate to Dominate: The Rise of the Chinese Techno-Security State, Tai Ming Cheung recalls that in 2017, Xi gave a speech in which he claimed that if China wanted to ascend to the top, it had to become an innovation powerhouse. Later that year, the Trump administration released a national security strategy in which a national security innovation base is identified as a necessary condition for American prosperity and survival. Both leaders shared the premise that national security requires them to articulate and develop national systems of innovation.

One can view the confrontation between the U.S. and China not in terms of the Thucydides trap, but rather through the lens of the so-called Needham Question. Joseph Needham was a British biochemist and author of a 25-volume history of Chinese science. A key problem emerged from his research: how did it happen that China, which had been at the forefront of science and technology for most of human history, was outpaced by the West? The realization of its own technological weakness was painful, occurring during the first Opium War and after, during the so-called century of humiliation. Today, Chinas Communist elite not only wants to avoid technological vassalization, as Xi put it, but to take the lead in technological progress and astound the West, just as the West astounded China in the 19th century. Chinas national innovation system is meant to be a set of institutions and policies designed to achieve this goal.

According to Cheung, Xi Jinping is the opposite of Deng Xiaoping. In the 1970s, the latter reoriented the state away from military and security issues toward the challenge of capitalist reform. Large military budgets were slashed and priority was given to growth. While Xi doesnt ignore economic indicators, he places more significant focus on technological self-reliance, economic security and the modernization of the military. Cheung quotes him as saying that science and technology power determines changes in the balance of world political and economic power. Xi is a techno-nationalist.

Since the beginning of his rule, Xi has argued that China must accomplish a transition: abandon its role as an imitator and transform itself into an innovator. Only in this way will it become self-sufficient in critical fields of technology. In order to attain this, the Chinese leader is pushing military-civil fusion, a process of exchanging information, resources and capabilities between civilian and military actors in key technological and economic areas. Xi has had the MCF written into the CCP constitution as a national priority.

One of the most important initiatives under the MCF banner is the construction of a network of large laboratories where the civilian and military sides can exchange experiences and work together. Cheung notes, however, that the projects development has been painfully slow. He points to the bureaucratic opposition and fragmentation as the main reasons. Elsewhere, he suggests that the innovation strategy provides cover for the process of consolidating power and overcoming resistance embedded in the state apparatus.

There can be no doubt that, with Xi at the helm, China has ceased to adhere to Dengs maxim, Hide your capacities, bide your time. But did the transition from Deng's economic openness to the pursuit of self-reliance occur suddenly under Xi's leadership? Cheung does not provide compelling evidence for the thesis of a qualitative, rather than simply gradual, change under Xi regarding technological self-sufficiency and economic nationalism.

Over the past years of strained relations between Beijing and Washington, there has been much talk about a Sputnik moment. Opinions differed on what exactly constituted that moment. Some pointed to President Donald Trump, his tariffs and the pressure he applied to Huawei. Others mentioned year 2016, when AlphaGo, a program created by British machine learning experts, beat the best human player in the Chinese go game. Still others claim that the restrictions on semiconductor-related technologies imposed by President Joe Biden Biden last year unleashed an unprecedented mobilization in Beijing.

Cheungs book allows us to expand the historical perspective and propose a hypothesis of a certain continuity, more pronounced than the author himself would admit. In May 1999, the U.S. bombed the Chinese embassy in Belgrade. The CCP responded almost immediately by establishing the 995 New High-Technology Project. The ambition was to dramatically increase spending on weapon R&D. Influential military officials quoted by Cheung assert that without this program the modernization of Chinese military would not have happened on this scale or at this pace. We should be grateful to the Americans, admitted one general. Consider: Total Chinese defense R&D investment between 1999 and 2009 (the first ten years of the 995 program) was more than all the investment in defense R&D in the prior fifty years. Through this prism, Xi would represent only an intensification of a trend, not its source.

According to Dan Wang, a tech analyst who spent years in China, Xi has no real ambitions of a great reformer like Deng. The great power he has amassed in his hands is not matched by any great goal. Regarding critical technologies, his results have been disappointing so far: Beijing is fundamentally misunderstanding the chip industry if it can believe that semiconductors can be run as a national space project, estimates Wang.

As Wang notes, China has not made significant advances in either aviation or chips. However, it has managed to gain dominance in other sectors. In the case of EVs, huge subsidies have created a number of companies that are changing the landscape of the automotive market. As Fords CEO remarked, something monumental happened in our industry where China became the number one exporter of vehicles globally. It had always been the Germans and the Japanese. Chinese competition sows panic on the Old Continent; Carlos Tavares, the CEO of Stellantis, complained in an interview with Le Figaro that Europe rolled out the red carpet to Chinese companies, and now it will have to pay a steep price to compete with them. Westerners, says Tavares, are no longer comfortable with change, and rivalry with the Chinese will be Darwinian. How many will be able to adapt?, asks the CEO. China has accumulated unrivaled competence in the EV industry. According to Goldman analysts, in the PRC, construction of a EV factory takes 1/3 of the time required in other countries, while Chinese battery plants are 80 percent cheaper to build than American ones.

The PRC controls clean tech supply chains. Its grip on green technologies is so firm that many experts do not believe the EU can meet its green goals without maintaining close ties with China. When it comes to wind turbines, its share in the production of essential components amounts to over 70 percent, and in the case of solar panels, it reaches 80 percent. Beijings dominance in the area of rare earth metals particularly in refiningis equally strong. Of the 54 mineral commodities that the U.S. Geological Survey deems crucial to the country, America depends on China for as many as 35.

According to Jeffrey Ding, who studies the questions of technology and great power rivalry, it is not narrow technological developments that decide a countrys fate, but so-called general purpose technologies like the steam engine, electricity, or the computer. These are engines of growth that become engines of power, as the rise and fall of nations is ultimately determined by differences in economic growth.

Ding argues that the nature of the American innovation system presents a huge advantage to the U.S. Its decentralized character allows methods, information and technology to diffuse quickly and in multiple directions. At the turn of the 20th century, while other countries had the potential to rival the U.S. in terms of innovation, it was America that was able to diffuse interchangeable manufacturing methods because it had stronger connections between the frontier institutions, entrepreneurs, and engineers. This gave it the edge that would be key to the American Century. It was later to be repeated with computers, when Japanin contrast with the U.S.struggled to integrate them into its economy and institutional practice on a large scale. Ding contends that the U.S. should defend the status quo when it comes to its own innovation diffusion structures, as these are what will keep America ahead of the PRC, including in the coming age of artificial intelligence.

A number of voices suggest that the American innovation system needs a major overhaul. On the surface, it appears that there is a genuine renewal. The CHIPS and Science Act, the Infrastructure Act or the Inflation Reduction Act might imply that the state has regained its ability to mobilize resources and carry out industrial policy. Even if we set aside criticisms, such as the argument that the CHIPS Act doesnt represent a success for the semiconductor industry, but rather a win for the incumbents or simply a rescue plan for Intel, this whole effort may still lead to nowhere. As David Adler and William B. Bonvillian remark, Pathways necessary for diffusing new technologies and getting them to market are missing, including a lack of scale-up financing mechanisms. The vocational education system has withered as has the corporate lab system.

They argue that one of the main culprits was the fascination with information technologies. As a result, industrial policy was eventually dropped and deindustrialization allowed. Yet the problem boils down to more than a misguided notion of the countrys developmentAmerica is not going to turn into one big California, after all. The greatest barrier to technology implementation has become the lack of a technically trained workforce. This problem is also resurfacing in the context of the CHIPS Act. Of the 115,000 new jobs anticipated to be created in the semiconductor industry by 2030, 58 percent are projected to remain unfilled at the current graduation rate. If America wants to build an advanced manufacturing base at home, it must have the right workforce. Unfortunately, the vocational school system, as Adler and Bonvillian write, has collapsed. Without reconstructing it and reducing the gap between higher education and the manufacturing process, neither reindustrialization nor acceleration of technological progress to raise Americans standards of living will be attainable.

According to the head of the U.K. A.I. Foundation Model Taskforce, Ian Hogarth, we are entering an era when geopolitical relations will be transformed, if not destabilized by A.I. The U.S. innovation system will be put to a real test before our eyes with the advent of new general purpose technology.

Paul Scharres Four Battlegrounds: Power in the Age of Artificial Intelligence is a book about the age of A.I. nationalism. Its main premise is that an A.I. race between the U.S. and China is already underway. Translating this new technology into military or economic power will not be easy, but it will happen eventually. The author does not pay particular attention to the economic aspects of the race, such as the issue of explosive growth, which could give absolute economic advantage to one of the great powers. He is primarily interested in the effect that A.I. development will have on military power, as well as the conditions necessary for A.I. to produce such an effect.

Scharre does not define the essence of the A.I. race. Importantly, this arms race, as the economist Pradyumna Prasad explains, is indefinite: There is no clear finishing line, competitors relative position during the race is of utmost importance and there is always the possibility to reverse any win or loss. He believes that the most important edge that this technology could give in the future is the ability for total mobilization of economic resources in the event of war. What U.S. industry accomplished during World War II, could be achieved by A.I. in the near future. It could solve logistical problems, increase production efficiency, or magnify the effects of research and development.

Sharre describes in a very accessible manner the battlegrounds on which countries are competing in the age of A.I. nationalism. One of them is data. The claim that China is the Saudi Arabia of data is inaccurate; China has a wealth of a certain type of data, but lacks others. While their facial recognition systems can be trained on extensive material, it is unlikely to be used to train fighter jets.

The hardware aspect is overlooked by the public, and perhaps constitutes the most complex issue of the A.I. race, as it involves the globally dispersed semiconductor supply chains, where the Dutch ASML and Taiwans TSMC play such an outsized role.

Algorithms are not a scarce resource. People are, Scharre notes. Chinas potential in this regard is growing steadily: Between 2009 and 2019, the number of AI researchers increased twelvefold. PRC researchers are expected to surpass Americans in 2025 when it comes to ranking in the top 1 percent of most cited articles. However, it is still America that attracts the crme de la crme of A.I. engineers.

The non-obvious battleground in the A.I. race is institutions. Its not just the structure of diffusion of tacit knowledge and innovation that matters, but also their internal dynamism. Scharre cites the example of an American company that worked with the government in the field of A.I. and whose owner concluded that he had no other choice but to let his start-up be acquired by a larger corporation that had entire teams dedicated to government contract compliance. As Scharre writes, the fact that an AI start-up felt it needed to be acquired by a major defense contractor in order to succeed is a major problem for the DoD. The ossified bureaucracy needs reform if the U.S. doesnt want to lose its lead in the technological arms race.

Regulation is another facet of the competition for the best institutions. The European Union has thrown itself with ferocity into the race to regulate AI. It is widely believed that Brussels wants to be the quickest and the toughest, because that way, as in the case of GDPR, it will succeed in imposing AI standards around the world. That being said, its hard to imagine that the U.S., China and especially Big Tech will just stand by and watch. As Brookings analysts have convincingly shown, the so-called Brussels effect has little chance of working in the case of A.I. The regulations designed by the EU are so strict that supposedly none of the so-called foundation models (like GPT4 or LLaMa) meet them. One can risk the thesis that E.U. is well aware that the Brussels effect will not work this time around.The A.I. Act is, as I see it, the Maastricht of A.I.all member states will now be bound by an even tighter web of laws regulating a technology that will perhaps shape our future to a greater extent than the Internet.

The A.I. Act will also establish in each country a new agency associated with European A.I. regulation. In doing so, it will create a new interest group, the A.I. bureaucracy, which may align more closely with the E.U.s interest than with those of member states. A.I. is not an opportunity for Brussels to assert its regulatory power in the world, but to deepen the European integration.

The PRC has opted to be less strict. The Chinese state crafted a set of fairly rigid regulations and put them up for debate by companies working in the industry. In consequence, the initially tough restrictions have been loosened. Still, the most relaxed approach prevails in the United States. This attitudewhich one analyst described as laissez-faire and learnamounts to keeping a vigilant eye on AI, while acknowledging that lawmakers do not yet know enough about the technology to formulate regulations.

Washington's decision not to burden A.I. with excessive regulations demonstrates its confidence in Silicon Valley. Not only are U.S. companies spending more on A.I. than the government, they are also capturing all the available talent: In 2020, 70 percent of PhDs in A.I. were hired by the private sector. On the other hand, while China can count on the complete loyalty of its companies, the U.S. governments relationship with Big Tech is problematic.

Scharre reminds us how 3,000 Google employees protested their employers cooperation with the DoD. If support for American national security was controversial for Google, cooperation with the CCP on the Dragonfly project (it involved a search engine for the Chinese market that would meet Communist censorship requirements) was not. Googles CEO defended the cooperation with China to the very end.

Similar protests over cooperation with the U.S. government have also broken out at Amazon and Microsoft. However, there are companies in the field of AI for which national security is important, like Palantir. Its CEO, Alex Karp, rightly noted that entrepreneurs from Silicon Valley charge themselves with constructing vast technical empires but decline to offer support to the state whose protections and underlying social fabric have provided the necessary conditions for their ascent. They would do well to understand that debt, even if it remains unpaid. Ultimately, the U.S. will have to carve out a narrow path between Big Techs dominance and the governments heavy-handed grip, treating A.I.-building companies as it used to treat Big Oil: supporting them where their interest align with the advancement of the American national interest, while also curbing harmful monopolistic behavior. Nevertheless, the shift to techno-nationalism will be too painful for some techno-cosmopolitan incumbents will be replaced by companies that are not ashamed to work with the U.S. government.

The A.I. race carries obvious but asymmetric risks. It is to be feared that if A.I. gets out of hand, it will happen in the PRC. As Bill Drexel and Hannah Kelley write, Little accountability for mistakes means that business owners tend to play fast and loose with safety, as evidenced by Chinas grisly history of industrial accidents. In those cases where accidents have come to light and been met with public outragelike with the toxic toothpaste incident, the poisoned infant formula, or the collision of two high-speed trainsit has done nothing to improve public safety.

Achieving mutual restraint in this arms race seems to be no easy feat. If the hype proves true, the advantage in AI could dwarf all other factors, becoming, in the parlance of Jake Sullivan, a force multiplier. But even if AI does not translate into a military or economic revolution, it will significantly change both war and the economy, and no great power will miss the opportunity to gain an upper hand in those areas. Scharre mentions that there is a lively discussion in China about AI restraint. However, he cites a U.S. general arguing that those who talk the loudest about restraint are the ones who implement it the least. The trust gap seems unbridgeable.

To ensure that America never has to consider the Needham question in relation to its own development trajectory, it needs to ask itself a different one. What made it rise to first place in terms of technological progress? Many books have been written on the subject and complex arguments constructed to account for American exceptionalism in this regard.Still, there is a perspective that seems neglected. It reveals something about the American character. One study shows that CEOs born in frontier counties are more supportive of innovation than others, and their companies create more and higher quality patents than their competitors, building a culture striving after technological breakthroughs. Fixing the innovation system alone wont be sufficient: the technological frontier could still vanish from the horizon. The frontier spirit has shaped America and expressed a certain moral type, but it is not eternal and may fade away. The fifth battleground is to preserve and cultivate it.

See more here:
The Vanishing Frontier - The American Conservative

Alphabet: The complete guide to Google’s parent company – Android Police

Rebranding gives businesses an image refresh and a competitive edge in an era of dynamic markets. Google underwent it in 2015, creating the Alphabet we now know as its parent company. Its birth has made it possible for its divisions to operate independently. Each remains a part of the company while handling projects beyond the internet search engine, advertising, or making new Google Pixel phones.

Google remains the Google you know but falls under Alphabet as a subsidiary and the largest shareholder. It runs alongside Waymo, Calico, and other companies in its diverse portfolio. Learn more about Alphabet, who runs it, and other information in this post.

Alphabet is a multinational technology company that Larry Page and Sergey Brin created on October 2, 2015. Page and Brin are Google's co-founders and restructured the popular technology company to expand and diversify their operations.

Alphabet and Google aren't the same. The former became the parent company, and the latter is now a subsidiary of it. Google shares have also converted to Alphabet stock and retain their ticker symbols as GOOG (Class C shares without voting rights) and GOOGL (Class A common stock) on the NASDAQ stock exchange and other platforms.

According to Page in an open letter, the name Alphabet fits the rebranding as it's a "collection of letters that represent language, one of humanity's most important innovations, and is the core of how we index with Google search." It also reflects in the website address as abc.xyz.

Nothing about how you use Google's products and services has changed. The Workspace apps, YouTube, and Maps, among others, remain intact. The difference is in the corporate structure. Current and future subsidiaries under Alphabet have more autonomy to chase separate goals and enter new markets.

Also, Alphabet began generating financial reports in three segments on a quarterly basis. They report the profit and losses for Google Services, Google Cloud, and Other Bets. Before that, there were reports for only Google and Other Bets. The segments operate as follows:

Google's overhaul makes the new company more accountable. Its introduction of the above divisions allows investors to monitor the financial performance of core services and startup projects. It also isolates the risk attached to each subsidiary, where one could fail or face roadblocks without affecting the others.

Different shareholders and investors own Alphabet as it's a public-traded company. Google's co-founders, Larry Page and Sergey Brin, hold its Class B shares. It gives them 10 times more voting rights, even though they own a small percentage of the total shares. Class A shares have only one vote per share, while Class C has none.

Also, B shares aren't public. Hence, they don't exist on stock exchanges and allow the founders and CEOs to control the company's direction and decision-making. In terms of executive positions, Sergey Brin was Google's President from the company's founding date in 1998 until 2019.

Meanwhile, Page acted as the CEO three times. First, from the founding date until 2001, then from 2011 to 2015. That same year, he became Alphabet's CEO and handed his position at Google over to Product Chief Sundar Pichai. Both co-founders stepped down from their positions in 2019 but retained board membership and are still major shareholders.

Pichai is now Google and Alphabet's CEO. He was the brain behind ChromeOS and played a pivotal role in Nest's acquisition, among other achievements.

Under the Alphabet umbrella are Google and Other Bets. Other Bets are companies still in their early or experimental stages and operate independently of the core internet services. Below are some of the subsidiaries Alphabet oversees:

Google's transformation story embraces change and progress, an effort that may continue to bring financial success and tackle public scrutiny concerning user data privacy. Post-restructuring, the new company has raised mixed reactions from supporters and critics. Some say that it may be setting unrealistic goals in the name of pursuing new horizons.

One includes Google Fiber and Webpass, two services meant to deliver fast internet and phone privileges to you via a physical line. Already, the company has had to pause operations in numerous cities and made massive layoffs. Speculations are abuzz about low demand and financial setbacks. But innovation is risky, and only time will tell if Alphabet's moonshot projects succeed.

Read more:
Alphabet: The complete guide to Google's parent company - Android Police

How AI and ML Can Drive Sustainable Revenue Growth by Waleed … – Digital Journal

PRESS RELEASE

Published October 6, 2023

The impact of AI and ML on modern business environments is more than fascinating; it's critical in today's hyper-connected world. While AI and ML have far-reaching practical applications, their greatest disruptive influence may be in business revenue. In this piece, I'll break out why artificial intelligence and machine learning aren't just "nice to have" but a "must have" for any company serious about long-term success.

The Importance of AI and ML in Generating Revenue

In today's data-driven and rapidly evolving environment, tried and true money-generation techniques are no longer sufficient. McKinsey reports that companies using AI in their operations boost revenue by 20% and save expenses by 30%.ChatGPT, GitHub Copilot, Stable Diffusion, and other generative AI applications have captivated global interest due to their widespread accessibility and user-friendly interfaces.

Unlike AlphaGo, which had a more specialized focus, these tools offer almost anyone the ability to communicate, create, and engage in uncanny discussions with a user. It's not merely a wave of the future; it's today's currency.

Practical Applications of AI and ML in Revenue Generation

Several revenue-generating uses for AI and ML exist:

These features may be added to your company model incrementally over time rather than all at once.

Challenges to Adoption and Solutions

The apparent complexity of the technology, concerns over data privacy, and the early expense of deployment are the most prevalent obstacles to AI/ML adoption. Based on my expertise in Turn-Key Design and Systems Integration, I would suggest a staged adoption, beginning with smaller projects to show rapid wins and ROI. In addition, working with other IT companies helps soften the change and save startup expenses.

Increasing Productivity While Lowering Expenses

AI/ML is a tool for improving the efficiency of an organization in addition to helping it make more money. With machine learning, everyday tasks are taken care of by computers. This frees up people to work on more complicated tasks and reduces technical debt. The production can also benefit from AI's ability to simplify back-end activities.

Tendencies and Prospects for the Future

The mutually beneficial connection between AI and ML and their earning potential will deepen as technology advances. Companies that don't change with the times will likely fail in today's fiercely competitive economy.

Final Thoughts

No company that wants to expand its income in a scalable and sustainable way can afford to ignore artificial intelligence and machine learning. It's not a matter of 'if,' but 'when,' AI/ML will become essential to your company's operations.

Who is Waleed Nasir?

Throughout his career, visionary builder and technology specialist Waleed Nasir has launched over a hundred platforms and led countless system deployments and workflow integrations. Dr. Waleed has extensive technical expertise in AI and ML and practical experience building and expanding technology companies. Notable examples of his work include the COVID-19 Crisis Management System, the Paycheck Protection Plan's Programmatic Loan Forgiveness System, and the Emergency Rent Relief Administration System. His wide-ranging skillset includes not just Turn-Key Design but also Process Automation and High-Performance Infrastructure, making him an industry leader in areas beyond only technological innovation. Currently, Dr. Waleed is working with Qult Technologies as the CPO, leading the company to new fronts.

Additional Resources

For those interested in diving deeper into this subject, I recommend:

Media Contact Company Name: qult.ai Contact Person: Hassan Tariq Malik Email: Send Email Country: United Kingdom Website: https://www.qult.ai/about-us/

Excerpt from:
How AI and ML Can Drive Sustainable Revenue Growth by Waleed ... - Digital Journal

The better the AI gets, the harder it is to ignore – BSA bureau

Hong Kong based Insilico Medicine, a pioneer in AI-based drug discovery, has made significant strides in recent years. Two of their candidates have reached clinical trials, with INS018-055 leading the pack as the first AI-discovered drug designed by generative AI to enter phase 2 clinical trials for idiopathic pulmonary fibrosis (IPF). Back in 2014, when the company began, AI for drug discovery was relatively unheard of, but now it's an indispensable part of the drug discovery process. Insilico's partnerships with major pharmaceutical firms like Janssen underscore the growing importance of AI in this field. Dr Alex Zhavoronkov, Founder and CEO of Insilico Medicine, sheds light on the industry's evolving response to AI in drug discovery, partnerships, regulatory reforms etc. and also shares the company's future plans.

Insilico Medicine has garnered attention for its innovative utilisation of artificial intelligence (AI) in drug discovery. Could you provide insights into how the industry's response to AI-based drug discovery has evolved since your inception in 2014?

In the early days, when I presented at conferences on how generative AI technology could be applied to chemistry, there was a lot of scepticism. I had discovered through my research that generative adversarial networks (GANs) combined with deep reinforcement learning (the same AI learning strategy used in AlphaGo) could generate novel molecules that could be used to treat disease. Since that time, AI drug discovery has undergone enormous acceleration, fueled both by advances in AI technology and in massive stores of data. While there are still no AI-designed drugs on the market, there are a number of companies with these drugs in advanced clinical trials, including our own lead drug for idiopathic pulmonary fibrosis, the drug with an AI-discovered target and designed by generative AI now in Phase II trials with patients.

Although the pharma industry has moved cautiously, the inherent risks in drug discovery (99 per cent of the drugs fail in the early discovery phase and 90 per cent of the drugs fail in clinical trials) and the validation of AI developed drugs to reach advanced trials, means that pharma companies are more actively pursuing partnerships and developing their own internal AI programmes. We have major partnerships with Exelixis, Sanofi and Fosun Pharma to develop new therapies, for instance.

Recently, your two candidates INS018_055, ISM8207 have entered phase II and phase I respectively. Can you share the significance of reaching these stages in the drug development process, and what key milestones do you hope to achieve during these trials?

To our knowledge, Insilicos lead drug for IPF INS018-055 - is the first drug for an AI-discovered target and designed by generative AI to reach Phase 2 clinical trials with patients.

AI was used in every stage of the process. Insilico Medicine used its AI target-discovery engine, https://insilico.com/pandaomics, to process large amounts of data including omics data samples, compounds and biologics, patents, grants, clinical trials, and publications to discover a new target (called Target X) relevant for a broad range of fibrosis indications. We then used this newly discovered target as the basis for the design of a potentially first-in-class novel small molecule inhibitor using its generative AI drug design platform, Chemistry42.

Insilicos molecule INS018_055 - demonstrated highly promising results in multiple preclinical studies including in vitro biological studies, pharmacokinetic, and safety studies. The compound improved myofibroblast activation, a contributor to the development of fibrosis, with a novel mechanism and was shown to have potential relevance in a broad range of fibrotic indications, not just IPF.

The current phase II study is a randomised, double-blind, placebo-controlled trial to assess the safety, tolerability, pharmacokinetics and preliminary efficacy of 12-week oral INS018_055 dosage in subjects with IPF divided into 4 parallel cohorts. To further evaluate the candidate in wider populations, the company plans to recruit 60 subjects with IPF at about 40 sites in both the US and China.

If our phase IIa study is successful, the drug will then go to phase IIb with a larger cohort. This is also the stage where our primary objective would be to determine whether there is significant response to the drug. The drug will go on to be evaluated in a much larger group of patients typically hundreds in phase III studies to confirm safety and effectiveness before it can be approved by the FDA as a new treatment for patients with that condition. We expect to have results from the current phase II trials next year.

Advancing ISM8207 is also significant both because it is the first clinical milestone reached in our partnership with Fosun, and also because it is the first of our cancer drugs to advance to the clinic, and cancer represents the largest disease category in Insilicos pipeline. This drug is a novel QPCTL inhibitor, designed to treat advanced malignant tumours, and works by blocking the tumour cells dont eat me signal. We entered into phase I clinical trials to assess the drugs safety in healthy volunteers in July 2023.

You have had quite successful partnerships with Exelixis, Fosun etc. Can you provide insights into Insilicos approach to forming strategic partnerships? How do you approach deal making?

We have the advantage of being able to produce and advance new, high quality small molecules that have been optimised to treat diseases much more quickly than traditional drug discovery methods. Thats because our generative AI system can optimise across 30 parameters at once based on desired criteria when generating molecules, rather than the traditional method of screening libraries to find a potential compound, and then working to optimise it for each desired property in a linear fashion. As we speed up the drug discovery process on these high-quality molecules we now have 31 in our pipeline we look to find partners who have specific disease expertise and clinical experience to advance these molecules into later stage clinical studies, and, we hope, to market where they can begin helping patients.

Our most recent partnership with Exelixis is a perfect example. We just announced an exclusive global licence agreement with Exelixis with $80 million upfront granting Exelixis the right to develop and commercialise ISM3091, an AI designed cancer drug and potentially best-in-class small molecule inhibitor of USP1 that received IND approval from the FDA in April 2023. This company is expert in cancer and cancer drug development and discovery, and has an expert drug hunting team. Because its an extremely innovative company, they already have substantial revenue coming from best-in-class cancer therapeutics and they are strengthening this pipeline and making bets on innovative cancer drugs.

If we were to look at one of your AI-designed drugs versus a traditionally designed drug candidate, is there a telltale signature?

Our AI-designed drugs will often have a novel structure or work via a novel mechanism compared to existing drugs. By optimising across these 30 different parameters to design molecules with just the right structure and properties to provide the best likelihood of treatment without toxicity and minimal side effects, we are essentially designing ideal new drug-like molecules from scratch. There may be other drugs that are designed to act on those same targets, but ours are optimised through structure or mechanism to be most efficacious, first-in-class, or best-in-class.

Until recently perhaps, big pharma was somewhat sceptical or resistant to AI. What has been responsible for this growing appetite to embrace AI as a fundamental part of the drug discovery process?

There are a number of reasons pharma is now embracing AI. Traditional drug discovery is an incredibly slow and expensive process that fails in clinical trials 90 per cent of the time. AI improves all three of those roadblocks improving speed, lowering cost, and optimising molecules to have the greatest likelihood of clinical trial success. Our AI engine known as PandaOmics can sift through trillions of data points quickly to identify new targets for disease that humans might not find. Then, our generative AI Chemistry42 platform can design brand-new molecules that are optimised to interact with those targets without causing adverse effects, scoring them based on which are likely to work the best. Finally, using our InClinico tool, we can predict how these drugs will likely fare in clinical trials to reduce the time and money lost on failed trials.

There is also now significant validation that this method of developing new drugs is producing very high quality new drugs for hard-to-treat diseases and even diseases that were considered undruggable. And a number of these AI-designed drugs are now in later stage clinical trials.

Finally, the technology is itself progressing and improving with additional use and data via reinforcement learning and expert human feedback. The better the AI gets, the harder it is to ignore.

How sceptical are regulatory bodies towards AI-driven drug discovery? How are regulations evolving to support such developments?

Data privacy and protection are critical to any businesses utilising AI, as is compliance with all international laws and regulations. I expect that these measures will become more stringent in coming years and they are essential to building and maintaining public trust. Insilico Medicine uses only publicly available data and employs privacy by design and by default. We facilitate security of our systems by thorough security analysis on each phase of development. All Insilico data hubs are contained in Amazon Web Services (AWS) or Microsoft Azure cloud.

In addition, there are several checks and balances in place to ensure continuous data integrity, protection and privacy. For example, clients data is not used in any internal environments of the platform, and a firewall is separated for the clients access to the platform versus everyone elses access. All data is encrypted, and data privacy is managed according to Insilico Medicines privacy policy.

What does the future hold for Insilico over the next few years?

Were eager to see our clinical stage programmes progress, and the continued advancement of our lead drug for IPF. Its a terrible, chronic condition with a very poor prognosis and patients are in desperate need of new treatment options.

I also hope that our latest deal with Exelixis marks a trend of pharma companies partnering earlier in the drug development process with highly optimised AI-designed molecules as we continue to expand our pipeline, so that we can truly accelerate the process of delivering new treatments to patients in need.

We will also continue to expand the capabilities of our end-to-end generative AI platform, through new data, reinforcement learning, and expert human feedback; and augment those capabilities with our AI-powered robotics lab as well as incorporating the latest technological tools into our platform, including AlphaFold and quantum computing both of which weve published papers on.

Ayesha Siddiqui

Read the original here:
The better the AI gets, the harder it is to ignore - BSA bureau