Not Rahm Emanuel
It’s great to see Argentina push for greater innovation despite the many challenges faced by the public sector. I think the open data initiative could be particularly impactful, but it also presents some challenges. As mentioned in the article, open data allows private companies to innovate and develop products based without significant cost to the government itself, since private companies can use the data to create applications that appeal to consumers directly. With that said, there are two risks that accompany making data widely available.
1. There could be issues with privacy, even if the data is only presented in aggregate or anonymized. I’m not sure what the typical Argentinian stance is on data privacy, but I know that people in many other countries are concerned with making their data public, even if it is impossible to identify individuals in that data.
2. Open data can drive government accountability because external parties are able to examine the government’s performance across a variety of metrics, but this can generate “noise” if the public focuses on high visibility, low value add, issues instead of those that are harder to discern from the data, but have a much more significant impact.
While I’m not convinced that the practical applications of this company are astronomical (insert groan for dad joke here) due to issues with quality standards and and technical complexities of manufacturing in space, I am also concerned with the funding challenges briefly mentioned in the article. Since the vast majority of space exploration is publicly funded, I imagine that most potential customers are highly risk averse and would not be willing to try this technology until the cost savings have been proven, as a significant loss in public funds could bring additional scrutiny on agencies like NASA.
On a more hopeful note, this article brought two moonshot (literally) applications to mind. Both are a long way away off and possible not feasible at all:
1. A large scale 3D printer on moon could potentially construct structures that would not be feasible given the difficulty getting large premade structures out that deep in space. Additive manufacturing is already being used in the construction of homes on Earth, and this technology, coupled with the proper communication technology, could allow humans to build on the moon prior to even having a person physically present (assuming the 3D printer can be preprogrammed or operated remotely).
2. I’m curious if the technology used for the Archinaut could also be applied to deep space probes. I’m not sure if this is currently a limiting factor, but presumably a robot could be installed on the deep space probe to detect and make small repairs whenever required during the probes journey. This would allow our probes to go further with fewer concerns of mechanical failure.
I think HBSStudent0918makes a really good point that this style of open innovation may not go far enough, though I think two small tweaks in their model could drive even greater innovation. First, rather than simply letting individuals pitch and vote on ideas, they could create an opportunity for the community to suggest incremental improvements to the top ideas, iterating to create an even more unique and interesting product. Second, users who have had their idea selected should be marked as a “power user” of some kind, which would give them additional resources and a platform to showcase their own idea and comment on others. These users would then be able to generate fresh ideas and move the market forward in ways that simple crowd sourcing cannot, as it plays on what is currently popular, but not what may become popular if consumers are educated properly.
I think the second question “How do we ensure equitable access to city resources?” is the one that concerns me most when reading this article. It’s clear that machine learning can help cities make more efficient decisions by guiding resources towards high density areas, however, I worry that this could create a cycle of divestment for other neighborhoods. For example, if a commercial corridor struggles to drive traffic, this algorithm might make fewer bikes available in that neighborhood, or it would be less likely to repave their roads. As the infrastructure and services become increasingly dilapidated, traffic will continue to decrease thus continuing the cycle. This would also be compounded by private developers who would be incentivized (even more than they already are) to build in areas likely to receive significant public sector infrastructure investment.
It’s fascinating, and frankly a bit horrifying to see how this model can be so biased against minority groups. In my post, I highlighted a similar program in Chicago that uses machine learning to predict where violent crimes may occur. The data used in that case can easily be biased as it is based primarily on reports from police officers who may be biased in how they write reports or how they current police in general. I would have hoped that collecting data that is not inherently biased (e.g., a surveillance camera) would remove some bias from the predictive algorithm, but unfortunately that does not appear to be the case. I’m wondering if that is because the data that initially created the algorithm had a disproportionate number of minorities who had committed crimes, or if there is another confounding factor here. I am curious if the Chinese police force has a way to manually adjust for these biases, or if they are simply following the algorithm. Given the fact that most algorithms continue to learn over time, a small bias against minorities initially, will compound over time.
I think that this technology can be tremendously impactful if deployed the right way, and agree with the many concerned mentioned around racial (and other biases). One question i’d like to see PredPol tackle with its partner cities is how to increase data and remove bias in data collection. If the data gathered is all based on police officer observations and crimes reported by civilians, there may be a whole host of crimes that are committed and never reported. By concentrating police efforts in areas with frequent crime reports, the algorithm may bias simply towards areas where residents are more likely to report. To combat this challenge, I would like to see some technology deployed in concert with PredPol, such as ShotSpotter, (https://www.shotspotter.com/), which has been deployed in Chicago to aid in their police analytics efforts.
I think this is an outstanding application of machine learning, and I do not believe that students would be particularly concerned about data privacy, as younger individuals tend to be much more comfortable and trusting of technology. My larger concern with this application is how universities will decide to employ a retention solution on their own campus. Building an algorithm in house is incredibly costly, requiring both data infrastructure and programmers, while looking for a solution externally is risky as well. There are many companies in this space from start-ups like Ellucian and Hobsons to larger entrenched players like Blackboard and Moodle who are expanding from the LMS space, but it is incredibly hard for universities to determine whose algorithm is best given the black box nature of machine learning algorithm. While any proven solution is likely better than nothing at all, it would be a shame for a university to be dissuaded from making an investment because they do not want to turn their students’ data over to an external provider they don’t fully understand.
Similar to AR, I am also highly skeptical of the proposition that MOOCs will be able to replace traditional universities for all of the reasons mentioned above. However, I see real value in MOOCs and believe that they are not actually competing with Ivy League Universities, but instead will develop increasing relevance for three different communities.
1) Displaced workers: As automation grows more prevalent, many workers will need to re-skill in order to find new jobs. These individuals are unlikely to go back to school, but may be able to join an adjacent industry or function if they receive some targeted instruction where their work experience and MOOC courses are relevant.
2) Community colleges: Many community colleges focus on highly technical skills targeted to increase their students employment prospects in a specific industry. Employers in these fields do not always require an associates degree, which would allow individuals to take MOOC courses to build those skills in lieu of community college.
3) Supplemental training for job seekers: At the college and graduate school levels, a students courses will rarely prepare them for the specific nuances of a job interview (e.g., consulting case interviews or financial modeling exercises). MOOC courses could be designed specifically for these interviews to act as a supplement to the students existing education. While not formally a MOOC, this is somewhat similar to how Trainingthestreet has been successful.
I agree the other comments that civic duty alone is not enough to encourage public sector innovation, but I actually think that NYC has an opportunity to make a compelling economic case without spending public funds. NYC could take the BigApps contest to the next level and instead of crowd sourcing ideas that they then select to scale and purchase themselves, they could make all of the licensed datasets mentioned publicly available at all times. This would give programmers the data they need to develop B2C apps that create social value for city residents, AND would give them a monetization model turning loose the open innovation and removing government from the selection process.
Similar to NYC Writer, I’m curious how significant the improvement in outcomes will be for custom 3D implants compared to current orthopedic products. Since cost savings appear unlikely in the short term, demonstrating this difference will be critical in order to increase adoption. This may partially be due to consumers/physicians comfort with current products, increasing barriers to switch unless outcomes are significantly higher, but I believe that an even more significant barrier will be insurance companies. Insurers typically put constraints on what they’re willing to reimburse, and they will see little incentive to reimburse the new products unless the improvement in outcome is more significant than the increase in costs.
This technology offers a lot of promise across the construction industry, but I’m curious where it is likely to have an impact first. With the ability to build small structures quickly it seems like a natural fit for affordable housing on the surface, but those developments are made economically feasible by using low cost materials and constructing significant density (a large structure with many small apartments) both of which may not be possible given the restrictions of additive manufacturing. Instead, it will need to be coupled with modular construction processes, where sections of a home (or even full apartments) are built offsite and then assembled to make the final structure, almost like building a Lego set.