Biden’s Executive Order on AI- deep dive

[Music] Hello everyone, this is Rohit with another episode of the identity navigator. Uh, thank you for listening in. I really appreciate it.

U, and I do have a good episode lined up for you if I say so myself. Today we'll be talking about uh the Biden's executive order on artificial intelligence and according to me it's it's truly a master class irrespective of your political affiliations but before we deep dive into that uh some personal updates I have been a little under the weather so you might hear me cuffing uh a little bit during the episodes I'll try to keep it at a minimal and I'll try to move the microphone away from my face as much as possible. possible.

So, thank you for understanding. Uh, second thing is I did have to travel for work a little bit. I I think I caught the bug there.

So, uh it is what it is. Uh we had our department's end of the year town hall. Uh that went well I would say.

So excited about new year uh work-wise and podcast wise and also have some things uh you know some some irons in the fire. So hopefully those would turn out good as well. Uh before we go ahead and start the podcast, a word from our sponsor.

And I got you again, didn't I? So I don't have a sponsor. Uh I am just a lonely guy sitting in my or on my kitchen table uh recording this podcast on the only piece of technology that is an Apple iPhone.

So as I don't have a sponsor which is quite funny and sad as well but uh that more to come on that later uh I will actually dedicate this episode to somebody. So nobody is sponsoring me but they cannot stop me from dedicating this uh and I dedicated this or I dedicate this podcast to Mr Steve Jobs. Um, not because I am some job fanatic or Apple fanboy, but only because of the fact that I am holding an Apple iPhone in my hand and that is the only thing I could think of.

So, uh, Mr Jobs, wherever you are, I hope you find peace or have found peace. And that's pretty much it. All right.

Uh so before I say something and put a foot in my mouth, let's start with something that I really know about and that is Biden's executive order on artificial intelligence. So before we do that uh let's let's have a philosophical view of what is future. Is it measured based on time or change?

So the argument can be made both ways. But for me future is when things change. It is related to time.

But the relationship is not linear. That means every 10 years you won't have the future worth 10 years. If that makes sense.

So let me try to uh give you an example. When you read the books or watch movies from early 20th century, 2020 or 2030 uh is predicted as a world in which we'll have flying cars or we'll have space travelers or total annihilation in some cases. But none of them has happened.

that hasn't happened not because the vision was wrong but that narrative was created based upon the advancements in the space of science and technology during the late 19th or early 20th century. So imagine this in almost 60 years or 65 years uh almost we went from not having flying airplanes or airplanes to landing on the moon. So just imagine from 1903 or someplace we had our first aircraft and then by 1969 we were on the moon.

So crazy 60 years. Have we been able to achieve that level of growth in the next 60 years? I don't think so.

But if we have kept with that pace, we will all be having flying jetpacks by now and working, right? So, uh during the late 20th century, this growth slowed down in terms of technological advancements. We did have some growth, but not the crazy growth that we witnessed in the late 19th or early 20th century.

Now for the 21st century according to me this race for AI supremacy will become a defining feature of our century. This will be marking a new era in technological advancements. Now this race is not just among leading tech giants like Google and Microsoft and Amazon or emerging startups but it is also a narrative of good versus bad actors in the digital world.

On one side, there are companies and researchers pushing the boundaries of AI to solve complex problems, enhance efficiency and unlock new possibilities in field like healthcare, environmental production and education. These positive application showcase AI as a force of good driving innovations and progress. On the other side, there is a darker aspect where malicious actors utilize AI for cyber attacks, deep fakes, uh, spreading misinformation, presenting significant eth ethical and security challenges.

So the contrast between these uses underscores a crucial responsibility for tech companies, governments, and regulatory bodies to ensure AI develops in a way that's safe, beneficial for society, and most importantly ethical. This AI trajectory makes the ongoing debate around AI ethics, governance, and regulations more critical than ever. As AI continues to evolve, the balance between harnessing its potential for positive impact and mitigating it risks remains a key focus for the global tech community.

Now for me I do not see it as a race between different product companies or tech companies or companies which are focused on users or good versus actors or bad actors. This is truly a race between chaos and bureaucracy. Why do I say that is because bad actors and startups I'm not comparing them.

I'm just saying they do not have a lot of processes in place. There is a chaos but only those will succeed who will focus on the right potential of the AI and good actors or big companies where there is a lot of bureaucracy will have to let go of some bureaucracy to win this race. So we will have to find or the winner would be that entity who will find the perfect balance between chaos and bureaucracy and I would say that the companies will control chaos will win the AI race.

Now when we speak about AI advancements we need to have guidelines in place. These the guidelines that I see bring that I see which are being put together today in multiple companies are mostly academic pretty basic or if I can say pretty much mostly CIA strategies in place which stands for uh cover your ass strategies. Now US has been the forefront of the technological advancements and the executive order by Biden's government is a masterclass in why US have been able to hold on to its technology supremacy over so many decades and and maybe centuries.

This is a master class this executive order in how to have the first mover advantage and encourage controlled chaos. So just some background uh President Biden issued a comprehensive executive order aimed at ensuring the safe, secure and trustworthy development and deployment of artificial intelligence or commonly called as AI technologies. This order is a part of Biden Harris administration's broadest strategy for responsible innovation.

Uh so let's let's let's dive in. Uh let's look at some of the key points from the executive order. Number one, new standards for AI safety and security.

So the order mandates developer developers of critical AI systems to share safety test results and other vital information with the US government. It requires notification to the federal government when training AI models that pose serious risk to national security, economic security or public health and safety. Number two, protecting against AI related risk.

The order includes measure to protect against the use of AI in engineering dangerous biological materials to prevent AI enabled fraud and dete deception and to establish a cyber security program to develop AI tools for identifying and fixing software vulnerabilities. Number three, promoting AI safety and security. So agencies like the National Institute of Standards and Technologies will set rigorous standards for AI safety testing before public release.

So there is also a focus on protecting critical infrastructure and addressing threats posed by as AI systems to it. Number four, which is for me the most important of them all, is protecting Americans privacy. The order calls for the development and use of privacy preserving tech technologies in AI or techniques in AI, strengthening privacy pre preserving research and evaluating how agencies collect and use commercially available information.

And I for one am very interested in understand to how are agencies collecting and using our commercially available information and and what model is being trained on that. Number five is advancing equity and civil rights. So guidance will be provided to prevent AI algorithms from uh propagating discrimination in various sectors and actions will be taken to address algorithm discrimination and ensure fairness in the criminal justice system.

Number six is supporting consumer patients and students. Uh so the order directs the development of resources to support educators deploying AI enabled tools and establishes safety programs in healthcare to report and remedy AI related harms or unsafe practices. Number seven is supporting workers.

principles and best practices would be developed to mitigate the harms of AI on workers focusing on job displacement, labor standards, workplace equity, health and safety and data collection. Uh this I guess would be uh the public favorite including myself because this is focusing on you know uh the effects on job displacements which I think is the number one concerns against AI usage for for most of us. Number eight is promoting innovation and competition.

So the order aims to catalyze AI research across the US, promote a fair and competitive AI ecosystem and expand the ability of highly skilled immigrants and non-immigrants with AI expertise to stay and work in the US. So I would not be surprised if uh US gives out additional highly skilled visa to the non-immigrants or the immigrants who are at the forefront of AI technologies. Number nine is advancing American leadership abroad.

So it includes action to collaborate on AI at international levels, establishing robust international framework for AI safe use and promoting safe and responsible AI development globally. And you know just just putting this executive order out, US has had such a big first mover advantage that will reap its benefits in decades to come if not more. Uh I did see that there was something from EU uh that was out today from European Union.

I haven't really had the time to deep dive into this but I will look into this. Uh and you know you can correct me and I hope that I'm not wrong but I'm expecting a lot of bureaucracy there. So the Biden's executive order is favored in terms of controlled chaos and I'm pretty sure EU would be favored in terms of or tilted in terms of bureaucracy towards bureaucracy but I I guess I could be wrong and I would be more than happy to see that I am wrong.

Number 10 is ensuring responsible government use of AI. So the order directs the issuance of guidance for agency's use of AI including standards to protect right and safety improve AI procurement and modernize federal AI infrastructure. So in summary when we looked at these 10 points uh President Biden's executive order encompasses a broad range of measures aimed at ensuring AI's development and use align with safety, security, privacy, equity and ethical standards while fostering innovations and maintaining American leadership in AI technology globally.

Right? So these were some of the key points. Now let's summarize the executive order's purpose before we deep dive into the executive order itself.

So the purpose is to establish a comprehensive approach to advance the development and deployment of AI in a manner that upholds American values of liberty, privacy and openness. It's privacy or privacy, I'm never sure. Its purpose is to ensure that AI systems are designed and used in a manner that is safe, secure and trustworthy.

Especially with AI being such a black box in in most of the cases, I think trustworthy is a really good things to look at here. So the core objectives of executive order includes ensuring the safety and security of AI technologists and technologies protecting the American public from potential risk associated with AI. Uh election fraud comes to mind right away.

Enhancing privacy and civil liberties as AI technologies are developed and deployed. Promoting AI that is equitable and does not reinforce existing biases. Supporting workers and education system in adapting to the changes brought by AI.

Fostering public trust in AI technologies, which for me would be a big challenge, maintaining America's technology leadership in AI on the global stage. And by setting out these goals, the executive order seeks to guide both the federal government and private sector in responsible AI innovation, focusing on public trust and setting global standards for the development and use of AI. So just so far I think this is pretty comprehensive.

I think whoever has drafted it is very knowledgeable. uh it's not just a bunch of buzzwords but really the things that we should really be focusing on u and for me it is uh pretty well drafted though I am speaking to some of the other AI experts uh in the industry next week. So hopefully they would listen to this podcast and they would have they can open my eyes like they can come up with something that I haven't really thought about.

So I'm I'm happy to be educated on those matters. So now let's discuss the broader context of AI in government policies and national strategies. And this encompasses several key areas.

So, national security and defense, economic growth and competitiveness, health care and public health, education and workforce development, you know, preparing the workforce for an AIdriven economy through STEM education and reskilling programs, ethics, governance and regulations, research and developments. So the strategies government strategies include significant investments in R&D to ensure or advance AI technologies infrastructure. So AI requires robust digital infrastructure right international collaboration and standards public sector adoption which is mostly the slowest sector to adopt a new technology especially the one that it does not know a ton about.

public safety and law enforcement. So AI tools would be deployed for public and safety purposes including crime prediction which is interesting, facial recognition and emergency response coordination, intellectual property and commercialization. So uh this means that strategies often include creating a supportive environment for the commercialization of AI including protecting intellectual properties and and supporting startups data privacy and cyber security.

The topic that fascinates me the most. And then in the context of United States the government has historically taken steps to ensure that it remains at the forefront of AI technology. This includes the creation of national initi in initiatives to coordinate AI research and policies across various federal agencies, partnership with private sectors and educational institutions to foster innovation and the establishment of ethical guidelines to regulate AI development and deployment.

Right? So, so this is like a broader context. Now let's look at the significance.

Obviously national strategy on AI is absolutely required because AI complex AI has complex challenges and opportunities. Uh and aligning various stakeholders in government, industry and academia towards common goal is a big task and I I think this executive order does a good job at that safety and security right. uh by setting standards for AI safety and security, we aim to prevent harm and unintended consequences that could arise from the development of AI systems.

You know, especially in critical areas about national security and public welfare. But the key to note here is unintended consequences. There are things which we will not be able to think about and we will have to have the boundaries in place or the checks in place for those controlling those unintended consequences.

So we are talking about significance right ethical framework is obviously required economic impact. So this will have a transformative e economic impact and we need to be able to harness its potential for growth while mitigating job displacement and ensuring fair labor practice. Hm fair labor practice that is going to be interesting right uh global leadership so again US aims to maintain and extend its leadership uh at a global stage and that's good regulatory clarity uh addressing discrimination international standards public trust inter agency collaboration innovation and competition and in a sense this this executive order is a strategic step to guide the responsible development and governance of AI, balancing the need to promote innovation and economic growth with the imperative to safeguard societal value and public trust.

Now let's deep dive into the order's focus on AI safety and security. Uh so the focus on AI safety and security is about establishing a set of standards and practices to ensure that AI systems are developed and deployed responsibly. Uh some of the key aspects are critical system review.

So the developers of crucial AI system may be required to submit safety test results and related information to the government. I think which is a good thing. Risk notification.

There is an obligation to notify federal agencies about AI systems in development that could pose significant risk to national security, economic stability or public health and safety. Cyber security measures. The order likely encompasses the development of AI tools that can identify and fix vulnerabilities protecting against AI enabled threat testing standards.

The order might mandate rigorous safety testing for AI systems, particularly those implemented in the public sector before they are deployed. So by focusing on these areas, the executive order aims to prevent potential misuses or accidents that could arise from powerful AI technologies, ensuring that they benefit society while minimizing risk. So now let's discuss the implication for developers and national security right so so let's start with developers first compliance and reporting right we've already spoken about developers of AI system uh especially those working on critical applications will need to comply with new safety and security standards uh this may include you know safety test results and risk assessment submission to federal agencies innovation versus regulation which is very important While we need to promote innovation, developers will have to navigate through more regulated environments which could influence the design, develop it, and testing phases of AI systems.

So imagine your project gets gets delayed because the federal agency hasn't responded yet. That would be fun. All right.

Transparency and accountability. Developers may be expected to document and explain their AI systems decision-m processes, which is a good thing. However, making developers document something might not be as easy as Mr Biden thinks.

Uh because this will on on the on a serious note, this would ensure transparency and accountability in AI operations. Cost of development. Now, adhering to stricter safety and security standards could increase the cost and time of development.

not good. It would increase the cost and time of development impacting go-to market strategies and budget and collaboration with government. Developers might need to work more closely with government agencies to ensure that AI systems do not pose unforeseen national security risks.

Now for national security, uh let's look at some of the implications, right? Advanced threat detection. The development of AI system could enhance national security capabilities in threat detection, intelligence analysis, and cyber defense.

Dual use concern. So, a technologies often have dual use potentials, meaning they can be used for both civilian and military applications, and this necessiates careful oversight to prevent misuse. International security dynamics The order may influence the global AI arms race, particularly in how the US and other nations approach the militarization of AI and related technologies.

I'm pretty sure this has already started. I don't want to sound like a conspiracy theorist, but I would be very surprised if the governments across the world are not looking into militarization of AI and tech and and related technologies. And if you are they are not they are missing out information sharing.

So there may be increased emphasis on information sharing between the government and private sector. Global norms and standards. So US sets his own standard for AI safety and security.

We will deep dive into the role of NIST a little bit more uh later in the podcast. Uh but this is good. So overall the executive order is likely to lead to a more structured and perhaps a stringent environment for AI development which I think is a good thing especially at the start with a focus on mitigating risk and enhancing the benefit of AI in a manner consistent with national security standards.

So let's explore some of the measures to safeguard AI in areas like bioengineering and cyber security. Let's start with bioengineering first. So ethical guidelines very important.

We need to establish strict ethical guidelines for the a use of AI in bioengineering to ensure it is used for beneficial purposes and does not lead to unethical manipulation of biological organisms. regulatory oversightes, safety protocols, again dual use research and transparency and reporting. In cyber security, we have security by design, promoting a security by design approach in the development of AI systems, which you should already be doing irrespective of AI or not.

uh vulnerability assessments. So conducting regular vulnerability assessments and penetration testing of AI systems to identify and mitigate security risk. A tip, you should already be doing both of these even for your non-AI systems for all critical systems.

Incident response plans, cyber security training, collaboration with cyber security experts, uh update and patch management. So all good things in cyber security, right? It's more about how not what.

So by implementing these measures, the risk associated with the use of AI in sensitive areas like bioengineering and cyber security can be mitigated, promoting the responsible and secure use of AI technologies. um let's discuss the challenges and hopefully some potential solutions. So let's start with challenges.

Let's let's look into the same areas that we have looked at. Let's start with bioengineering. So complexity of biological systems.

So these systems are complex and not fully understood making AI application in this field unpredictable. So the solution should be we need to increase investment in basic research to better understand biological systems before applying AI at scale because it is like unknown over unknown. It's like x + y is equal to 5.

We don't know x we don't know y. How do we solve for it? Dual use approach.

U research beneficials for society can also be misused for harmful purposes such as creating biological weapons. uh implementing strict controls and ethical oversight on research that can be dual used along with international treaties on biological research I think would go a long way and then one of the other challenge would be rapid pace of innovation. The pace of innovation may outstrip the ability of regulators and ethical frameworks to keep up and that is why we need that balance.

We need these checks and balances in place. Therefore, the solution is creating adaptive regulatory framework that could be updated as technology evolves because this technology will evolve. Uh challenges in let's let's look at some of the challenges and and solutions in cyber security.

So, cyber threats evolve rapidly, right? And AI systems can be both tools and targets in its landscape. So developing AI system that can adopt new threads dynamically and can work alongside human cyber security experts.

The next thing that I find really really useful not just terms of challenge in cyber security but also some of the privacy concerns that I have is the blackbox algorithms. Many AI systems are not fully transparent or in or interpretable which can make securing them difficult. So investing in explanable AI research to make algorithms more understanding and auditable.

And last but not the least challenge that I want to touch on is a scale and automation. AI can operate at a scale and speed that is challenging for human oversight. So establish multi-tered monitoring systems that include automated checks and balances as well as human oversight.

Easier said than done, right? Let's look at some of the general challenges. Uh regularity lag.

So you know there is often a lag between technological advancements and regulatory response. Uh we can close our eyes and say there would be not but we are pretty sure there would be right. So government should use a regulatory sandbox approach allowing for real world exper experimentation and controlled condition.

I think the government would have to do it if it wants to succeed in the race of AI supremacy. Global coordination. So AI and bioengineering advancements occur globally.

Uh this require international cooperation for effective safeguarding. So let's promote international dialogue and treaties to establish global standards and cooperate security cooperative security measures. Resource allocation.

So yeah, you know, resourcing is always a challenge now especially with public private partnership uh and then sharing the burden of investment in safety and security is another challenge and I think the common sense should prevail here. Talent gap is obviously you know you heard me saying about maybe a new type of visa might be introduced uh new courses in the university and maybe reskilling of some of our existing workforces but talent gap irrespective would exist. public trust uh it is pretty challenging right and the two biggest challenge area in terms of public trust that I see and even I have is is job displacement and loss of privacy right and this could be only be done or this could only be done via transparent communication about AI initiatives their benefits safety measures along with public engagement in policym with the state of our political system right now.

I am very very concerned about both of these challenges. And last but not the least is our thinking should be forwardthinking strategies, right? Effective policym, international cooperation, the safe and secure integration of AI into critical fields especially like bioengineering and cyber security.

This needs to be effectively managed. uh let's look at how agencies like NIST or National Institute of Standards and Technologies will play a crucial role right so some of the things that we expect from them would be setting standards testing and benchmarking research and development guidance for implementation collaboration with industry public policy support education outreach. So pretty self-explanatory points but NIST involvement or NIT's involvement would be central to ensuring that AI technologies are safe and secure aligning with the broader objective of executive order on AI.

So I think it's time for NIST to step up. I haven't really been the biggest fan of NIST. Obviously they put out great content but now they have to match the velocity as well.

So let's see how they do but uh we need an oversight or governing body on top of NIST as well to keep them accountable on on all these things that we are expecting from them. Now we need to also examine the balance between innovation and regulation. So this is a delicate one in in the context of AI.

On one hand, too much regulation can stifle creativity, slow down technology development, and potentially hinder a country's competitive advantage. And we are talking about countries, right? So that's a big deal.

On the other hand, insufficient regulations can lead to unchecked systems that may cause harm, violate privacy rights or perpetuate biases and discriminations, election fraud, anybody? And here is how the balance can be examined. So encourage innovation, use flexible regulatory framework, use regulatory sandboxes and provide incentives for innovation in the sense of tax incentives, grants uh and and similar things.

But also you need to ensure safety and trust. So safety standards, right? Privacy protection and ethical guidelines should be implemented, drafted uh and you governed.

And then finding the balance you know stakeholder engagement, iterative regulations, international cooperations, education and workforce development and risk assessment uh will all help us to find the balance. Now let's look into how does this order focus on preventing algorithmic discrimination. Right?

So fairness in AI it the order emphasizes the importance of fairness in AI systems which involves developing algorithms that do not inadvertently discriminate against certain group of people. uh this is particularly relevant in in areas such as criminal justice right like I don't have to explain this there could be so much of bias uh in criminal justice or hiring or lending or healthcare where biased AI could reinforce societal inequalities now you feel like you know you can take somebody's race out of that equation or model and things would be all right there's so much complexity to it it would be hard to remove these these biases diversity in design. So encouraging or mandating that AI systems can be designed and tested by diverse team is one approach to reduce bias.

Right? So this can help ensure that different perspective are considered and that systems are tested against a wide range of scenarios and data sets. Now this for me is goes you know back to the basics of what is considered diversity.

Are we looking at diversity in terms of gender or race or age uh or country of origin, but it's much more than that, right? I I was talking to my team the other day u and does a white male born and brought up in Midwest or let's say Wisconsin has the same values or style or whatever you want to call it as a white male brought and brought up in New York City, right? And and this becomes super complicated as you deep dive into it.

So I we were talking about you know if we hire two such people are we diverse right and that led to a discussion of how would we talk about diversity in AI technology so something for us to think about uh for preventing algorithm discriminations transparency and accountability audit and assessment and guidelines and standards right let's think about how does this order focus on preventing or protecting privacy. So data protection. So the order addresses the need for robust data protection measures to ensure that personal information used to train and run AI systems and not is not misused or leaked, privacy by design, anonymization techniques, right?

The use of data anonymization anonymization, right? uh not tokenization, not encryption but maybe like hashing and other privacy preserving techniques not even hashing because you will have to operate on that data. So the use of data anomalization and other privacy preserving techniques in AI could be promoted ensuring that AI can be trained and operated effectively without compromising individual privacy.

So I take it back hashing will not work here. we will have to do a true data anonymization, regulations on data usage and then public awareness. Uh last but not the least, let's include some perspective from civil right advocates.

Right? So inclusivity advocates would stress the importance of inclusivity in AI development teams. They would argue for diverse representations to argue that a wide range of experiences and perspective are considered to identify and mitigate biases in AI system.

Now this is a good statement on its face value but I would really like to understand what does inclusion and diversity means that should not just be a token gesture but it has to be interweaved into the fabric of the system and we will have to look beyond the conventional DEI initiatives which for me have been sometimes just just a two gesture I'm sorry to say that transparency uh accountability public engagement and then legal protection ction. There should be a push for strengthening legal protections against discriminations uh ensuring that existing civil rights legislations is adequately applied to and enforced in the digital [Music] realm. From the point of view of technologist, we'll have to think about bias mitigation, privacy enhancing technologies, ethical AI development, self-regulation and interdisciplinary collaboration.

So how do we get to a balancing perspective? So to balance these perspectives, policymakers would need to create forums for dialogue between civil right advocates, technologists and other stakeholders. Policy approach should involve core regulations where the government set broad objectives for preventing discrimination and protecting privacy and industry develop specific standards and practices to achieve those objectives.

continuous monitoring and iterate iterative policy development. So this is essential to respond to the evolving landscape of AI technologies and its societal impacts and by weaving together these perspective uh policy makers can develop a more comprehensive and nuanced approach to AI governance that seeks to maximize the benefit of AI while minimizing harms uh ensuring technology works for all segments of society. Let's look at how this order aims to support educators and healthcare providers and workers.

So for educators, this would be uh involving integration of AI into the curriculum, professional development. U this could include you know training programs, AI as a teaching aid and ethical education for healthcare providers. uh AI could be used in diagnostic and treatments.

Uh they would need to be provided training on AI tools, data privacy and security. HIPAN needs to be modified now. Uh research and collaboration and for workers.

So for all of you out there uh reskilling and upskilling, job creation, label stand labor standards and and protection uh and participation in AI governance. So we need to encourage worker participation in the governance of AI at the workplace ensuring that their voices are heard in decisions about how AI is used and managed. So for all of these sectors, let's look at u some of the long-term impacts.

So on education, you know, customized learning could be a thing. Adapting to each student's learning style, pace and preferences, access to education, uh we have done great there. I think with AI, we can do even more.

Uh teacher roles. So the educators may transition from traditional teaching roles to facilitators of learning as AI takes over some aspects of you know instruction assessment and then lifelong learning. AI may necessitate a shift towards lifelong learning.

Sucka as the rapid pace of technology changes will require continuous skill upgrade. So it will not just be my developer brothers learning a new language every 6 months and sisters. Developer brother and sisters.

My apologies. Long-term impact on healthcare. Uh improve patient outcomes.

Keeping my finger crossed. Uh because you know it could lead to more accurate diagnostics. Uh personalized treatment plan and improve patients outcome through advanced data analysis.

Operational efficiency, healthcare accessibility and then public health on the workforce. Uh job transformation. Many jobs will be transformed as AI automates routine task.

Hard as it may be, it is it is the future leading to a shift in the skills that worker needs to be successful. So it is not going to or according to me this would be a transition. Uh we will have to do some things differently.

That's that's pretty much what it would be. But it would also have new job creation, right? So it will create new industries and job categories most of which we cannot even think about right now.

which will lead to opportunities for economic growth and career development in emerging fields. Inequality concerns there is a risk that the benefit of AI could be unevenly distributed. I don't think this is a risk.

I think it's a high probability exorating income and opportunity disparities unless addressed by inclusive policies and training [Music] programs. Workplace dynamics. So AI can change the nature of work with more people working alongside AI systems or in jobs that are enhanced by AI.

So I'm curious if I have to work alongside AI system, do I still need to go to the office? Okay. Uh let's look at the order's role in fostering AI research and US maintaining its global leadership.

Right. So this order can lead to research funding and grants or increased funding for AI research. It can lead to collaboration between academia and industry, government-led initiatives and research in ethical AI.

All of which will actually foster AI research and for us to maintain global leadership. This order can actually position the United States as a standard bearer in AI. This is what I was talking about when I was talking about the first mover advantage by promoting the development of stand international standards that reflect American values and technical excellence.

It is like a land grab right now. Let us set the standards. International collaboration export of AI technologies.

So by maintaining a leading edge in AI, the US can be a major exporter of AI technologies and services influencing global markets and technology norms. influencing global governance, talent attraction and retention and competitiveness in AI. So ensuring that US remains competitive in AI development is likely a key goal of this order.

This includes investing in necessary infrastructure such as high performing computing and data resources to support advanced AI research. So how will we evaluate the order in the in the or this order in the long term? This we'll do this by effectively seeing whether it catalyzes in advancements in AI, whether it encourages responsible practices and navigates the international landscape of AI competition and cooperation.

It can be seen as a blueprint for ensuring that the US continues to harness AI's transformative potential while leading the global conversation on its terms. Right. Uh let's discuss a little bit about international implications and and partnerships.

So nothing new. Um I guess we have already discussed this a lot. It is just now in the context of you know global or different countries.

Global AI governance, international standards and norms, economical and geopolitical balances, collaboration in research and development, trade and commerce, ethical or ethics and human rights, defense and security and then cyber security. So overall the executive order on AI can be a catalyst for the US to shape the trajectory of AI development globally through leadership, partnership and active participations in international forums. The aim would be to ensure that AI grows as a force for good promoting prosperity, security and ethical practices worldwide.

So uh let's just recap the key points that uh we have in President Biden's executive order on AI. So we spoke about the purpose of the order to ensure the safe ethical and equitable development and deployment of AI aligning with American values and interest. We spoke about safety and security preventing discriminations privacy protection.

Uh we also looked into implications for developers in national security. We looked into how to safeguard bioengineering and cyber security industry. What the role of agencies both NIST and threeletter agencies would be.

Uh balancing innovation and regulation support for key sectors. We looked into you know education and healthcare and workers long-term impacts. uh again in these same sectors uh fostering research and leadership.

Uh we also looked into international implementations and the overall narrative of the executive order is to responsibly guide the development of AI that maximizes its benefits right and also positions the United States as a leader in the global AI [Music] landscape. So ending with some of my personal thoughts here uh on the future of AI policies and its impact on society. We need to be inclusive in policym.

It has to be created by voices from different technologist, ethicist, is ethicist a word? I hope so. Civil society uh and you know law or lawyers and from various other sectors affected by AI.

We need to have adaptive regulatory framework. The technology is changing at a pace. We need to have our regulatory framework go at the same pace if not more.

We need to focus on ethical AI. Global cooperation is a must more than anything else that we have ever resounded to. Maybe after uh nuclear treaties, these AI global cooperation would be an even bigger thing.

Economic and workforce consideration. Do not be capitalistic to the extent that it eliminates the majority of your workforce. Public engagement and trust, data privacy and security.

Something that I did not speak about but I've always thought about is how would it impact democracy? So how will the AI policy consider the impact of AI on democratic processes from combining misin or combating misinformation to ensuring that AI tools do not become instrument of surveillance and controls right even with even with information right it's pretty much like what do I want to feed you or what do I feed you in order for you to have a particular point of view And I don't think we are smart enough to actually fight against that. So this has to be fought at a large at a at a level larger than all of us.

We also need to invest in AI safety and security. So we also need to have an interdisciplinary approach that means just not focus on you know technology. It will definitely benefit from you know blending insights from technology and social sciences and law and even philosophy to create well-rounded robust policies.

This is where I started from a philosophical discussion of what is the future. So I would end by saying that the impact of AI policy on society is profound. Policies crafted today will lay the groundwork for how AI shapes our future.

From the economy to ethics, governments to national security, the challenge lies in crafting policies that encourages innovation and harness the potential of AI for public good while also safeguarding against risk and ensuring that the evolution of AI aligns with human values and societal goals. I hope you found this podcast beneficial. I'm going to be meeting some very learned individuals this week.

So hopefully I will come back and share any insight that they share with me. You can always reach out to me via LinkedIn uh or via my email the identitynavavigator@gmail. com.

Thank you all for listening and closing thoughts was I was looking at the different type of audiences that I have and the least amount of people listening to me are Zenz. So I asked my niece to how do I cater to Zenz audience and she said uh uncle Dulu is the new Salulu. So here I am either trying to either sounding too cool or too stupid by saying Delulu is the new Salulu.

So hopefully that would increase my Gen Z audience. But in the meanwhile, this is Rohit, your identity navigator.