>> please welcome courtney lang, vice president of policy at i.t.i. with congresswoman zoe lofgren. [applause] >> thank you, congresswoman, for joining us today at the intersect. and thank you to everyone for being here. for those of you who don't know, congresswoman lofgren represents the 18th district of california, is the ranking member on the house science, space and technology committee, and is also a senior member of the house judiciary committee. so really glad to have you here today. i have to say i'm happy to be sitting in this seat.

since i recently testified in front of the house energy and commerce committee and i have to say, it's nice to be giving the questions as opposed to receiving them. with that, let's get right into it. i wanted to start with an area that i know you have been focusing on quite a lot lately and maybe that's an area that, in the a.i. kind of era that we're living in, hasn't been talked about as much. but i really wanted to start with hearing your perspective on, as we're looking at other emerging technology areas, one of the things you've been looking at is quantum and why it's so important that the u.s. is prioritizing quantum computing and information science and what you've been thinking about looking at in that space. ms. lofgren: the science committee has taken the lead on this issue. we took the lead on 2018 on the quantum initiative that i think has yielded the results we had hoped for in terms of research and supporting efforts in the scientific world. one of the great things about

serving on the science committee as compared to the judiciary committee is that we are very bipartisan. the chairman, frank lucas, he's a conservative republican. i'm not. but that does not keep us from working together because he wants to get something done and so do i. so we are working on the re-authorization of the quantum initiative. we're about to begin a four-quarters discussion with the senate because we want to make sure that we're touching all the bases. democrats and republicans, house and senate, in that endeavor. obviously quantum computing is not quite there yet. although we are making tremendous strides with -- especially with the hybrids. but it's not just computing, it's also sensing and other uses of the technology and i do think it's important that we maintain our competitive edge there. when you think about all of the

things that are happening, sometimes i feel we're just on the door step. a.i., quantum, fusion, and you if we don't make some serious missteps in the next few years, our future is very bright on this planet if we continue the steps forward that we are trying to promote. >> thank you. you mentioned the quantum reauthorization act which iti was proud to support. what will that do for the u.s. quantum capabilities and how will that help to bolster those capabilities and maintain that competitive edge? rep. lofgren: we are trying to move beyond the truly scientific into some of the applied. nasa is a player, but they were always excluded from this, so we

are including them. we also want to make sure that u.s. scientists have the resources necessary, which is a big problem. i love the private sector. i love our private sector companies, but it is important that the scientists and academic researchers also have access so that they can make process -- progress. we had a hearing just yesterday on that subject. >> i wanted to ask you a little bit more about this in the context of ai in particular and certainly the national science foundation discussing the efforts on the pilot program which is going to be one important mechanism by which to advance some of these research and development activities as well as give access more broadly. we have been very supportive of the mission as well as some of

the legislative efforts like the create ai act to really do that. i'm wondering if you can share your thought on how we can make sure that there are federal resources available for this to really succeed moving forward. rep. lofgren: as i say, we had an excellent hearing in the science committee. we invited some nonmembers who are very interested in this. as well as don buyer who has had great interest in this. it is a resource issue as well as a workforce issue. the two are related. the barrier for entry is very high in terms of computing power. talking one of the leaders in openai expressed skepticism we would be able to provide the resources necessary to catch up.

i hope that is not the case because much as i love the private sector, ai should not be owned by three companies in the united states. it should be accessible to the academic community and to researchers and to scientists. and to start ups that matter.

not to mention the fusion program. courtney: we certainly hope to see that moving forward. this resource is going to be really integral to innovation and democratizing access and making sure multiple players in the ecosystem are able to harness this resource. we hope that those efforts are successful moving forward. i guess sticking on the topic of ai which will surprise nobody, it is a big topic of conversation right now and i

think it is important to remember that the science committee has been doing work on ai for over a decade now. certainly as the conversation progresses and as we moved to a place where there are different thoughts about how to best regulate and oversee the technology, i want you to share about how you and your colleagues are thinking about the areas you all are prioritizing. thinking about the right way to approach governance of this technology. rep. lofgren: we tasked them in number of years ago with doing a collaborative process and laying out a non-regulatory framework. i think they did a pretty good job of that. in the end, we are going to need some regulations. they have laid a foundation for

how we might proceed. there are disagreements on approaches. should this be risk-based? do you create one giant agency to do the whole thing. it is impossible for one agency to be an expert in every single thing. it is important that we know what we are doing before we do it. we have had some bipartisan informal briefings. so that we are better educated. one of the issues and this is a judiciary committee issue is the role of intellectual property. there are a lot of questions about that.

the product of ai and its connection to intellectual property. the registrar of copyrights is engaging in some investigation on that right now. i don't think in the end she will be the definitive voice. i think the courts are likely to be it. one of the things that she established and i think it is correct is that a creative product has to be created by a human. you can't copyright something that does not have a human touch. there is going to be a line. where is that line? she is going to work on that. the issue of ingestion is before the court right now. people are discussing whether 230 applies to this. i don't think that has any application to ai whatsoever. the whole theory was to protect free speech. that is not an issue for ai.

the real issue is what about humankind for the data that has been ingested? i also look for transparency. one of the things we are looking at in our ai discussions is greater transparency and better access to high-quality data so that ai could be improved, not just generative ai, but other aspects, where we have a lot of opportunities. people are afraid of misinformation. that is valid. there was recently a deepfake with taylor swift and president biden's voice was used as a fake. all of that is worthy of concern. on the other hand, the use of ai , let's just think about what google did. you would have a phd student

unfolding single protein in the human body to write a thesis on that. google unfolded all of them into weeks, posted them for everyone to have. it is going to charge the development of medicine and others, medical advances, tremendously. there are all kinds of wonderful opportunities that we need to keep in mind because there are a number of people on capital harrow -- hill whose hair is on fire and the sky is falling and that is the wrong approach as well. courtney: i really appreciate that perspective and it is important that we are recognizing how transformative and innovative the technology can be. while also recognizing the risks. we put out principles a couple of years ago on that so we welcome your guys' focus on that as well as some of the other challenging issues. copyright has also come up.

rep. lofgren: i would guess congress is not that great at legislating. we specialize in failing. the chances are in the near term other than what the science committee is going to produce which is not the big regulatory theme everyone is talking about it is the private sector efforts in the near term that will be addressing these issues. adobe is coming up with a tagging system that is very promising. others are also looking at that. i think we may see some real advancements. it is not going to eliminate the need for advancements, but the private sector will carry us a little bit farther and bad actors will tell us more of what we need to he worried about walk-on dress continues to be dysfunctional. courtney: there is a lot of work going on in the industry that people are thinking about including what you just

referenced. i think one other area i wanted to ask you about is you mentioned the role in pushing vista development in the ia -- ai risk management framework. that has been a foundational framework for organizations thinking through how did they manage ai risks through their lifecycle? but given where you said, i was curious if you could speak about the role that international standards play and standards in advancing the conversation on ai risk management and how that ties into the work folks are doing in the actual standards development space and implementing some of the ai executive orders or otherwise. rep. lofgren: i think this standard-setting is very important. we are not where we need to be, but if you can't measure it, it doesn't exist. we also need to work with the

developed world on international standards. i'm skeptical that we will be successful with some of our developed countries. china tends to go off on its own. i'm concerned that going off in a direction that may not be that positive. but i think if the united states can take the lead in standard-setting even if it is voluntary standard-setting, that will help some order into this thing. i think the science committee may play a role in that. we have some excellent scientists and technicians, both in the private sector, but also in government and i want to call them out for the excellent work they did pulling people together. we have an opportunity to leave and i think it would be a mistake not to do that. courtney: thank you. standards are going to be super important so i appreciate all you have done to really push

that forward. i think with that we are nearing the end of our time, so i really appreciate your joining us today and we will look forward to continuing to work with you moving forward. rep. lofgren: thank you, i would look forward to it and i will look forward to more work with the last bipartisan committee on capitol hill. [laughter] [applause] courtney: thank you. ♪ >> joining us now, the executive vice president of policy at idf, the deputy chief ai officer and director of cyber policy at the u.s. department of the treasury, director office of critical and emerging technology at the u.s. department of energy, and the

deputy chief technology officer at the fbi. >> well, good afternoon, everybody. thank you for joining us at the intersect. it is great to have you here today. we just heard from congresswoman lofgren and if you were at lunch, lots to talk about the importance of ai. the director speaking about the risk management framework read congresswoman lofgren said we are at an inflection for ai. if we are, it is because of the brass tacks of the real work agencies already have to do to implement ai and they are doing that because of the ai executive order and the guidance, the draft guidance from the white house to agencies that by my count spells out over 200 different things that agencies are going to be responsible for. that covers such a wide range of coverage thinking about how agencies are going to manage

risk not just for themselves, but the entities they are responsible for. it covers workforce and how you are going to fill capacity. it covers ai products and systems themselves and within the ecosystems that they exist. that is a lot. one of the first questions we all have is how our government agencies going to go about doing that? elena, i would like to start with you thinking about how are you prioritizing that? especially when we just saw agencies submitting initial inventories of ai and the department of energy submitted hundred really cool audits. there are cool things about the environment and energy. how are you thinking about balancing all that? >> that is a great question and we have a huge responsibility to deliver, both in terms of innovation, that is one big section of the executive order about how to train workforce, how to develop tools, foundation

tools, foundation models. we also have a big responsibility working with members of the interagency. this is providing safe and secure ai. model valuations. we were really pleased last week to submit to dhs our assessment of the sector risk management agency of the potential risks ai poses to the electricity grid. >> that is a really great point at the end because none of u.s. departments are in a vacuum. you have entities you are responsible for. brian, i know the same thing for treasury with the financial services sector. they call out the expectation that you are going to do that work read hopefully can share with us how that is going. >> as far as the executive order , we were assigned to produce a report on ai for the financial sector. we really did this as part of a

continuation of an effort we posted a year ago on cloud. we did that same model again. we started to talk to a bunch of different firms, financial firms, ai providers, other folks in the space to really kind of understand how ai is being used within the financial sector where a lot of the risk was about opportunities and how we think about this space going forward. i think some of the key illusions we had which are not finalized yet, ai has been used in the financial sector for many years. one firm has been using it may be up to 50 years and we heard some other small firms that are just starting to think about it. so these type of models aro is helpful within firms to understand the risks they are having and how to address that. but ai really fits in that risk governance model. they don't see it as a large change from what it is.

we always look back to this cyber framework. your biggest challenges always to identify how to make sure we are thinking about this. understanding how it may fit within their ecosystem going forward. as we are looking at this and continuing the conversation to understand what is being done in it? what are the challenges doing that? should we be working with industry to develop tools and techniques to harness the energy better and prevent bad things from happening? how do we work with all of that together? that public-private partnership and that engagement strategy has been really helpful. that is why we are excited to be here to listen to other folks. the speakers here, the people in the audience, understanding what your risks and challenges are. this organization has always been something we have looked to to thing about leadership. when you have comments and thoughts, please pass them on as we are trying to do our job and

develop policy to address these challenges. >> thanks. i want to come back to the governance points you are raising, but before you do that, it is not only brian r., there is a national security and law enforcement implication here as well. and a lot of what is in the guidance from the white house is about managing those risks to individuals and property and so on. i know the fbi has been doing a lot of work in that regard. hopefully you can give us some of the latest on that. >> certainly. i would like to maybe highlight two activities demonstrating our activities under the executive order 5960 that have again shown value.

the first activity is mapping risk to mitigation. the ai risk management framework. prior to that, we did a full on regulatory review on ethics and implementation. we grouped the risks and connected them to internal processes and technological control. a concrete example of a process control that we have identified. it is probably a common one. the fbi must ensure that our employees are fully trained and have the proper agency to assume the overall consequences of any decision that is made with naia or tool.

and so with that mapping as prerequisite, we then began the ai ethics council. this is patterned after a review board. for medical experimentations. we took great pains to ensure that the composition and expertise on that board reflects the bureau widely and the american population. the ai ethics counsel is complementary to the risk profile that we have identified. every risk case is evaluated on the merit of its risk and this allows us a forum to dive deeper into the nuances of ai. this is explicitly where we can

get into conversations about undesirable risk. in areas where ai in novel ways that traditional information technology systems do not. i feel that these two activities in concert with the totality of the fbi's ai response and i guess governance models have been hugely successful. they have been implemented by fbi employees and subject matter experts. i attribute a lot of that success to the conversations they have been having with their fellow government players, academia, and the mystery -- the industry more broadly. thank you. gordon: thanks, brian. i think a lot of what you said tied back to your points about the importance of government, having an ethics counsel and

risk profiles. one of the questions i have for you is how do you balance that with this desire for innovation for the mission people and you have mission responsibility to move forward as fast as possible, do not want to wait and say this technology will help us catch a bad guy or solve a financial rhyme or fix the climate and sustainability challenge tomorrow? we can't wait for the governance process. how are you balancing those engines? >> this is pretty exciting. as we are building out a new field here, i have been the deputy chief ai officer and if you saw in the new york times this weekend, the highest job across is the new chief ai officer which means there will be a lot of new people showing up in this space who have not been trained in it because there is no training at on what a chief ai officer is. that will be interesting to figure out how we build these components out with the ability

to understand the technology you are working on, understanding the mission you were working for , and making sure we tied that all together with the ethical issues another challenges we have. as the organization tries to deliver on its mission, you have to not slow it down, but jill build in the right efforts going forward. this goes into the whole education and hiring effort. i saw today dhs has 50 people who they are looking for in ai. i'm not sure how many within treasury we will be hiring, but there are not a lot of folks out there. how do you make sure you were pulling in the right folks with the right skills to address that. how do we make sure this does not get out of control? you know that ai and other things can always go into a direction not originally intended to read that is where governance comes into it. how do we make sure all the different components within treasury are aligned.

we have 13 different bureaus. that they are not coordinating so we are doing 13 times spend without the benefits we are getting for it. so understanding how to coordinate this is going to be really huge. how do we also work with our brethren in other departments and agencies like energy, fbi, and the like to understand how they are looking at it to align resources even better. working with trade and other folks like the folks here today to understand, how are you looking at it? how do we take best practices and incorporate those into what we are doing to be able to make sure we are doing this the right way? governance is hard. nobody necessarily likes it, but if it is done correctly, the benefit is huge. we know omb has a draft memo out there. you may have already commented

on it. if not, please do. it is still open. but making sure we are setting it up the right way. a lot of headaches for us in these positions, but the good headaches because hopefully we build out the effort so that the organization run better. gordon: i'm sure many of you in the audience help contribute to that guidance. there are lots of questions about intellectual property and ownership and how the government is going to manage all of that. that certainly is an active area where you know everybody is interested if i could follow up on that. the question i think the from members of the public, how do we provide continuous and real-time feedback? it feels as though it is moving so quickly, technology is changing so weakly.

>> part of my offices efforts for the last 22 years has been a public-private partnership. once the sector is doing it. we are highly dependent upon it. you can't run a bank without power. it doesn't work that way. how do we make sure we are engaged? send us comments. we love to read and learn. austin back to understand it the right way to do it? it is new in most places. hopefully some of you have ai officers out there that we start to engage with. what are your best practices so i don't make this a mistake. what do you do right. this dialogue is going to be critical. iti is critical to that to make

sure we are learning and understanding of the best way. over sharing is key to understand what is going on every of whatever we can do to get to a better spot is huge from our side. it is going to be a huge effort. not every sector will do this individually. we have to figure out how to do it best to get those economies of scale to build through and get to the right spot. gordon: just so everybody knows, i did not prompt brian to say come to iti with feedback. unsolicited, but we thank you. i will go to the other brian and then helena. you are all sort of one person at the moment. there is a huge workforce implication to all of this. the executive order and the memo

talk a lot about plans to build capacity, but it is early days. how do you think about expanding the workforce and bring in more people who don't just read an article about chatgpt or generative ai? what are you doing in the agency brian r.: we have programs to increase digital acumen the particular recipient is the university education fund. these types of opportunities really need to be fully explored in a more wholesome and comprehensive manner. ven back to the 2021 commission on ai which really

drove home the need i'm really hopeful that we are going to find some capacity in the future. gordon: you said your whole office is you and two fellows at the moment. [laughter] maybe congresswoman lofgren or senator blackburn will realize that some appropriations are in order for helping to address these, but what are you doing? that is not tenable for long term. >> i would say i'm really fortunate to work with a full slew of technical experts across all of our programs. of course, the over 50,000 scientists are truly a resource for the nation. they are not all ai experts, but they don't need to be.

we do have a tasking within the executive order. we are looking at building programs, expanding programs with the communities and the academic institution in their areas and seeing how we can ramp that up more. one thing i would add to all of this is people often associate us with clean energy specifically and climate. i think the thing that i would really like to share is that we have so much enabling infrastructure. to solve today and the future. both from the computing that we have. it is second-most energy efficient. it is something we have tried to couple through our partnerships.

i think this is a resource we should be able to use in aid of the nation. gordon: how does industry find out about those things. you visited some companies about those things. what do we do collectively to close that down. helena: i feel like it is my personal responsibilities to be leveraging this for the greater community. d.o.e. is very deeply interested in these areas and the energy efficiency that will need to underpin a lot of this future work. in terms of how people find out about it, we are also tasked to

develop partnerships with industry in the academia to build some of these tools, to leverage our computing facilities to be able to push the frontiers of this field further. gordon: i'm sure people here would have lots of interest to help doing that. people from oak ridge would have to come sell to individual agent, we have a computer. do you guys have any uses for it i know there have to be better ways. brian, i would like to ask you the same question. you are doing this external facing research right now and making import, but what can industry be doing to find out what is going on treasury engaging with them? brian p.: we have the federal private-public partnership. we encourage every financial institution to engage with that and work that partnership model

to understand what the challenges are, what they are seeing with the solutions they may have to be able to help. you can always call us directly. we are doing our 150 day report. we are getting best information from folks within the sector. what we have to keep doing is listening more. solutions are going to come out. how do we make that an environment that is comfortable with that discussion? the report we did on cloud, we did not cite anybody with it. we heard a lot of negative things that they had addressed.

that is our goal, to really make that a safe environment to share that information. one thing to add generally is i have to get smarter on this too. things that you find her helpful that you use within your day to be able to make sure you understand this. we get smarter on that stuff. some are highly scientific papers. whatever those are. the more conversations we can have that identify the things to help us get smart across the board is going to be helpful. i don't have time to go through every scientific publication, but i know a lot of people in this scouter reading stuff and say, this is much better than this other one. send that over to iti to be able to say, here is something we found that is helpful so we start that conversation and the more we can build that in and use you as a sounding board to get to us the right information, hopefully we will be in a better spot. hopefully, we get all invited

back next year and we will be able to say, this is what we learned over the last year. is this the right stuff? are these the challenges you are seeing? and then be able to focus on that conversation. to have that conversation over and over again. here are the things that are helpful. that feedback is going to be critical. how do we use the opportunities to get better? gordon: thanks, brian. we only have a few seconds left, but one of the questions that that all bags and it is great to hear that and if you are going to disclose report results, you are welcome to come back and do that here, but one of the challenges for us and for member companies and industry's treasury has their approach for that engagement and outreach and energy has there were roach.

-- their approach. helena, how do you think industry should be thinking about interacting across all those agencies as well? [laughter] helena: certainly, associations are one way. if there are things we should be considering, we are happy to hear it. gordon: you heard it there. helena willie for phone number and email address another way out. [laughter] and feel free to contact her. we are out of time and we will yield the stage to the next panel. thank you. [applause] ♪

>> please welcome iti's executive vice president the policy for a discussion with senator marsha blackburn. [applause] >> i'm honored to be joined on the stage with senator marsha blackburn who represents the state of tennessee. she is in a powerhouse set of committees in the senate including the finance committee, the commerce committee, the judiciary committee, and the veterans affairs committee. i had an opportunity to testify against -- at a committee and we talked about how to increase transparency in artificial intelligence systems. senator blackburn has been a leader in pro-innovation policy for all technology and in

particular for artificial intelligence and for quantum computing. i want to start there with ai. there has been vigorous activity on the hill. there have been a series of ai insights, discussions, and the two committees you serve on have done a lot of due diligence in the area of artificial intelligence. based on the discussions you have had, at this point, what do you think it is congress can do to promote a pro-innovation artificial intelligence agenda for the united states so it continues to lead the world? sen. blackburn: that is exactly what we should be doing is realizing that congress will create the environment for jobs growth and innovation. we need to be certain we do not legislate to a technology, but focus more on the utilization and use of that technology. that is the mindset that we should go into this with.

and i have been so pleased that we have been able to have some hearings, judiciary committee has been very active. commerce committee. we even had a veterans affairs hearing. rob: i missed that one. sen. blackburn: on the utilization of ai and what it could do. we have a relationship of the good, the bad, the ugly when it comes to ai. then you look at entertainment and we have issues around name, image, and likeness. my hope is we focus on the environment, we put the guardrails in place and we

realize that if we are not allowing innovation, china is. rob: we talk about having a tech neutral framework. sen. blackburn: that's right. rob: you touched on these many industries, tennessee is a leader in transportation, health care for many years, increasingly in telecommunications. of course the wonderful artists that do the performing and the songwriting in tennessee will certainly be able to use artificial intelligence. i want to bring you to another technology that we touched on briefly in the program today and that is quantum information sciences and quantum computing. i know that oak ridge national lab is a leader and you have taken a keen interest in this. what is the promise for

tennessee down the road? sen. blackburn: yes indeed, oak ridge has the world's fastest computer. we like to take a little victory lap on that one and celebrate it. what we have seen is one of our cities, chattanooga, tennessee, that is in close proximity to oak ridge has focused on quantum. chattanooga is a very innovative city and they focused on broadband acceleration, fixed wireless broadband, and enabling innovators to be there. they actually have a city quantum initiative. i did a quantum roundtable this summer and it was actually fascinating because of that. we have a couple of bills we are pushing in the commerce committee.

the quantum initiative needs to be reauthorized. in order to move forward and senator ray luján and i have a quantum sandbox bill that would allow that innovation that needs to take place. to be able to move from r&d into commercialization. how we build these quantum systems so that there can be a more widespread commercial use. likewise, there is a second bill we are working on that deals with testbeds and making sure we have the testbeds that will be in a classified setting or a commercial setting.

we want to make sure we do our part in creating that wraparound and that environment so that we can keep that innovation here. so that we can know that there will be a support system to elevate that technology. rob: you are being prescriptive and a narrow way to make the use -- best use of taxpayer dollars when you set up these public-private partnerships and framework for innovation to make sure that is transferred in a smart way with the private sector leading a lot of those initiatives for success down the road. sen. blackburn: that's right and we want to make certain that they are doing that. when we look at what is happening in the military space, we need to look at those partnerships and how we do

security clearances space forces going to be an area on quantum and ai. when you talk about next generation warfare and generative ai and some of the applications that can be there, be utilized, so yes, it is important for us to think forward on that. rob: we are glad you are there doing it. tax policy. tech companies to get what they need to go and scaling up invest more in r&d from their base revenues than any other sector of the economy.

we need good tax policy. last week, the house passed the bill that includes immediate r&d expensing. how would you prognosticate the future in the area of tax relief this year? sen. blackburn: we anticipate this piece of legislation that moved from the house and senate will come to us at finance committee. everyone agrees it is vitally important to do for people that are innovative in this space. we have such a short window and a lifecycle for many of these

technology applications. you need to be generous in allowing them these rmd credits. and this expensing, this immediate expensing, because they are going to be in to the next iteration and we need to keep that. it doesn't matter if it is quantum or ai, that structure needs to be there. on ai, let me say this. some of you are focused on what will happen with the entertainment industry and senator kunz and i filed a bill with klobuchar and tillis adorning us -- joining us in it and it is called no fakes to put in place the protections for name, image, likeness, and voice when it comes to ai to be certain we protect the creative community.

you've got chairman cantwell working on some ai bills. you will see other components that are going to come from judiciary that will focus on protecting intellectual property. so that you have the right to benefit from the work that you do. getting the patent and copyright those are all in our bucket of bills. rob: sounds like that will be a very active legislative agenda. sen. blackburn: we want to hear from you. rob: well as i mentioned, our time is at an end, i just wanted to ask the audience to join me in thanking you for being such a

leader for innovation and championing american innovation, as well as joining us today. sen. blackburn: thank you it is my pleasure. thanks a lot. [applause] ♪

's remarks are about 45 minutes.