Responsibly deploying AI in the immigration process

Some good practical suggestions. While AI has the potential for greater consistency in decision-making, great care needs to be taken in development, testing and implementation to avoid bias and to identify cases where decisions need to be reviewed:

In April, the federal government sent a request for information to industry to determine where artificial intelligence (AI) could be used in the immigration system for legal research, prediction and trend analysis. The type of AI to be employed here is machine learning: developing algorithms through analysis of wide swaths of data to make predictions within a particular context. The current backlog of immigration applications leaves much room for solutions that could improve the efficiency of case processing, but Canadians should be concerned about the vulnerability of the groups targeted in this pilot project and how the use of these technologies might lead to human rights violations.

An algorithmic mistake that holds up a bank loan is frustrating enough, but in immigration screening a miscalculation could have devastating consequences. The potential for error is especially concerning because of the nature of the two application categories the government has selected for the pilot project: requests for consideration on humanitarian and compassionate grounds, and applications for Pre-Removal Risk Assessment. In the former category of cases, officials consider an applicant’s connections with Canada and the best interests of any children involved. In the latter category, a decision must be made about the danger that would confront the applicant if they were returned to their home country. In some of these cases, assessing whether someone holds political opinions for which they would be persecuted could be a crucial component. Given how challenging it is for current algorithmic methods to extract meaning and intent from human statements, it is unlikely that AI could be trusted to make such a judgment reliably. An error here could lead to someone being sent back to imprisonment or torture.

Moreover, if an inadequately designed algorithm results in decisions that infringe upon rights or amplify discrimination, people in these categories could have less capacity than other applicants to respond with a legal challenge. They may face financial constraints if they’re fleeing a dangerous regime, as well as cultural and language barriers.

An algorithmic mistake that holds up a bank loan is frustrating enough, but in immigration screening a miscalculation could have devastating consequences.

Because of the complexity of these decisions and the stakes involved, the government must think carefully about which parts of the screening process can be automated. Decision-makers need to take extreme care to ensure that machine learning techniques are employed ethically and with respect for human rights. We have several recommendations for how this can be done.

First, we suggest that the federal government take some best practices from the European Union’s General Data Protection Regulation (GDPR). The GDPR has expanded individual rights with regard to the collection and processing of personal data. Article 22 guarantees the right to challenge the automated decisions of algorithms, including the right to have a human review the decision. The Canadian government should consider a similar expansion of rights for individuals whose immigration applications are decided by, or informed by, the use of automated methods. In addition, it must ensure that the vulnerable groups being targeted are able to exercise those rights.

Second, the government must think carefully about what kinds of transparency are needed, for whom, and how greater transparency might create new risks. The immigration process is already complex and opaque, and with added automation, it may become more difficult to verify that these important decisions are being made in fair and thorough ways. The government’s request for information asks for input from industry on ensuring sufficient transparency so that AI decisions can be audited. In the context of immigration screening, we argue that a spectrum of transparency is needed because there are multiple parties with different interests and rights to information.

If the government were to reveal to everyone exactly how these algorithms work, there could be adverse consequences. A fully transparent AI decision process would open doors for people who want to exploit the system, including human traffickers. They could game the algorithm, for example, by observing the keywords and phrases that the AI system flags as markers of acceptability and inserting those words into immigration applications. Job seekers already do something similar, by using keywords strategically to get a resumé in front of human eyes. One possible mechanism for oversight in the case of immigration would be a neutral regulatory body that would be given the full details of how the algorithm operates but would reveal only case-specific details to the applicants and partial details to other relevant stakeholders.

Finally, the government needs to get broader input when designing this proposed use of AI. Requesting solutions from industry alone will deliver only part of the story. The government should also draw on expertise from the country’s three leading AI research institutes in Edmonton, Montreal and Toronto, as well as two new ones focused specifically on AI ethics: the University of Toronto’s Ethics of AI Lab and the Montreal AI Ethics Institute. Another group whose input should be included is the immigration applicants themselves. Developers and policy-makers have a responsibility to understand the context for which they are developing solutions. By bringing these perspectives into their design process, they can help bridge empathy gaps. An example of how users’ first-hand knowledge of a process can yield helpful tools is the recently launched chatbot Destin, which was designed by immigrants to help guide applicants through the Canadian immigration process.

The application of AI to immigration screening is promising: applications could be processed faster, with less human bias and at lower cost. But care must be taken with implementation. Canada has been taking a considered and strategic approach to the use of AI, as evidenced by the Pan-Canadian Artificial Intelligence Strategy, a major investment by the federal government that includes a focus on developing global thought leadership on the ethical and societal implications of advances in AI. We encourage the government to continue to pursue this thoughtful approach and an emphasis on human rights to guide the use of AI in immigration.

Source: Responsibly deploying AI in the immigration process

What the previous government learned about birth tourism: My article in Policy Options

Excerpt are my concluding observations:

All this being said, the number of births by foreign mothers should be monitored. Statistics Canada numbers may not present an accurate picture. The number of births to foreign women in Richmond was reported as 394 in 2016-17, greater than the 313 that Statistics Canada reported for the whole country for 2016 (see table above). The Richmond numbers showed a steady increase from 2010, compared with the flatter trend in national numbers. Statistics Canada and IRCC need to work with provincial health ministries to ensure more reliable and consistent data.

More focused measures need to be considered to reduce or contain birth tourism. Options include making it more expensive by increasing the deposit that mothers pay hospitals; making suspected birth tourism grounds for visa refusal; and banning or regulating “birth tourism hotels,” places catering to pregnant foreign women and the consultants who help make the related arrangements.

These concrete actions would be a more proportionate response to the concerns raised by politicians and their constituents, and one that should be pursued by any government to improve the integrity of the citizenship program and address public concerns about fraud and abuse.

It is also important that the motivation behind discussion and debates on birthright citizenship not be labelled as racist, xenophobic or anti-immigrant. The fundamental issue remains fraud and misrepresentation, not discrimination.

What the previous government learned about birth tourism

Citizenship: What the Census Tells Us

Please find below the link to the Policy Options article I did with respect to citizenship and the related deck that I will present later this week at the Metropolis Conference in Calgary (hence will not be blogging for the rest of the week).

What the census tells us about citizenship

This analysis uses Census data to examine naturalization rates with respect to gender, age, education, immigration period and category, labour force status and median income.

“White settler revisionism” threatens Métis-Crown reconciliation

The complexities of identity, “peoplehood” and rights:

The 2016 census revealed explosive growth in the self-identified Métis population in Canada. The 51.2 percent growth of self-identified Métis from 2006 to 2016 easily surpassed the growth of First Nations and Inuit populations.

The growth is spread unevenly across Canada. Notably, the Métis population skyrocketed in areas where no historic Métis communities were located. Recently published research by scholars Adam Gaudry and Darryl Leroux reveals that the self-identified Métis populations in Nova Scotia and New Brunswick surged by 900 and 450 percent, respectively.

Clearly, demography alone cannot explain the trend. According to Gaudry and Leroux, people in eastern Canada are claiming Métis identity based on Indigenous ancestry revealed through genealogy. They call the practice of reimagining racial identity based on the existence of long-ago Indigenous ancestors as “white settler revisionism.” Many of those claiming to be Métis base their revisionist identity on the mistaken assumption that a mix of European and Indigenous ancestors is a sufficient basis to claim a Métis identity.

Far from being a harmless phenomenon, white settler revisionism systematically devalues Métis peoplehood by disregarding the process that led to the ethnogenesis of the Métis Nation.

The Métis Nation arose in the specific period after European contact and prior to European control of the specific geographical area referred to as the Métis homeland. The Métis homeland is a vast area now covered by the provinces of Manitoba, Saskatchewan and Alberta, and stretches into portions of Ontario, British Columbia and the Northwest Territories, as well as the northernmost plains of the United States.

The mass usurpation of Métis identity also has the potential to derail efforts at reconciliation between Indigenous people and the federal government.

Indeed, widespread assertion of Métis identity has the potential to stymie future policy frameworks. The Daniels decision, which held that Métis people are to be considered “Indians” for the purposes of section 91(24) of the Constitution Act, 1867, raises the possibility that the federal government will be required to provide more programs and services to Métis people.

Policy-makers must have a clear idea of the scope and distribution of a population requiring government support and engagement. The proliferation of dubious claims of Métis identity in disparate geographic areas poses serious obstacles to policy objectives.

The unscrupulous practices of organizations claiming to represent Métis people cannot be ignored. The Métis Federation of Canada, for example, does not require their members to prove Métis ancestry. Neither does the Bras d’Or Lake Métis Nation. To join these organizations, prospective members must simply demonstrate that they have an Indigenous ancestor. This ancestor can be Métis, Inuit or First Nations.

But the Métis Nation is not a simple conglomeration of ancestors with mixed ancestry. These organizations are creating chaos by convincing millions of Canadians that they are Métis, regardless of a lack of ancestral connection to the Métis Nation.

The Métis National Council and its provincial organizations, on the other hand, have meticulously crafted citizenship criteria that require concrete proof of Métis ancestry. In short, applicants must self-identify as Métis and demonstrate that they have an ancestral connection to the Métis Nation that arose in the historic Métis homeland.

Canada must intervene to ensure that the Métis National Council is not lost among an avalanche of illegitimate organizations. The federal government has begun this process by providing funding in its 2017 budget to the Métis National Council and its affiliated organizations. The money is going toward governance capacity and to support the council’s membership registry.

But more action is needed. Ottawa must affirm the Métis National Council’s resolution declaring that “there is only one Métis Nation, and that the geographic homeland of the Métis Nation is the historic Northwest which entered into Confederation in 1870 through the negotiations of the Métis Provisional Government led by President Louis Riel.” Only a clear and unequivocal statement will have the intended effect of silencing specious claims to Métis identity.

Additionally, policy-makers in Ottawa must understand that enabling the federal incorporation of dubious organizations like the Métis Federation of Canada could be harmful to reconciliation efforts with the Métis people.

Finally, Canada should provide funding to the Métis National Council so it can judicially intervene in response to illegitimate legal claims to Métis rights. A number of these claims have arisen in recent decades. Most recently, unsuccessful Métis rights claimants in New Brunswick sought leave to appeal to the Supreme Court of Canada a decision by the New Brunswick Court of Appeal that upheld the lower-court ruling that no historic Métis community existed in the province.

The Métis Nation and the federal government are on the cusp of achieving lasting agreements that will facilitate reconciliation and a just resolution to generations of conflict. But the proliferation of white settler revisionism and the mass usurpation of Métis identity threaten those prospects. The federal government must take seriously the threat posed to the Métis Nation by white settler revisionism, and continue to enact policy reforms to support the Métis National Council.

via “White settler revisionism” threatens Métis-Crown reconciliation

ICYMI: Is Canada’s population too small? My review of Doug Saunders’ Maximum Canada 

For those interested, my take in Policy Options on Doug Saunders’ Maximum Canada.

Source: Is Canada’s population too small?

Acknowledging that Canada’s hate groups exist

Good assessment and commentary on Canada’s right-wing hate groups by Amarnath Amarasingam and Ryan Scrivens:

The results of a three-year national study published last year — involving interviews with Canadian law enforcement officials, community activists and current and former right-wing adherents, triangulated with analyses of open source intelligence — suggested that the foundations of hatred are complex and multifaceted, grounded in both individual and social conditions that strengthen and weaken the movement. Understanding these conditions, from a policy perspective, will provide us with a starting point to counter right-wing extremism in Canada.

Some of the factors that strengthen the hate movement include: Canada’s history of racism; a political climate of intolerance that arises from time to time (for example, during Quebec’s Charter of Values debate); media (mis)representations of particular minority groups (for example, Muslims depicted as terrorists after 9/11); and a weak law enforcement response to hate groups. In many respects, it is Canada’s national political and social climate that enables bigotry and hatred to exist in the country, a climate that provides right-wing extremists with a backdrop against which they can recruit new members and spread their radical beliefs.

On the other hand, factors that weaken or destabilize this already unstable movement include a general lack of ideological commitment in hate groups and infighting and transiency within them, as well as strong and visible law enforcement response in certain localities, resilient communities and the presence of an antiracist movement. In other words, hate groups in Canada are generally unorganized and lack the ability to strategize and sustain themselves, and law enforcement officials who put pressure on these groups are generally successful at dismantling them. This is particularly the case when antiracist movements and communities work closely with law enforcement and share intelligence on hate groups or their adherents.

It is difficult to know exactly what the policy response in Canada should be to right-wing extremism, but, in general terms, we need to acknowledge that hate groups do exist in Canada and that they do pose a threat to the safety and security of our communities. Coming to terms with this is a first step in developing policy initiatives to resist the radical right. Discussions about counterterrorism or counterextremism policy cannot focus solely on jihadism. The Trump administration has rightly come under criticism for its decision to cut federal funding for organizations that are fighting right-wing violence, such as Life After Hate.

Second, we must directly exploit the strengths and weaknesses that are inherent in hate groups and their environments in order to disrupt their growth and sustainability, through an approach that includes individuals from different sectors of society. In other words, policy initiatives must include the voices of key stakeholders who have unique insight into right-wing extremism, including law enforcement officials, community activists and former right-wing extremists.

Information about Canadian hate groups is fragmented. Law enforcement officials, for example, may have one important piece of information about a particular hate group while community activists or former extremists may have another. Policy initiatives must bring these stakeholders together to develop effective responses to the threat from the radical right in Canada. We mustn’t look at the violence in Europe and the United States and complacently conclude that we are somehow immune. We are not.

Source: Acknowledging that Canada’s hate groups exist

The pendulum swing on government service fees: C-44 Service Fees Act proposal

The recommendations from my  brief to the House and Senate Finance Committees and the related article in Policy Options, based in part of how the previous government was able to increase fees twice within a year with minimal scrutiny:

While it is unlikely that the Commons and Senate finance committees will undertake a serious review of the proposed “Service Fees Act,” the following suggestions would reduce the possibility for potential government abuse of fee setting:

  • Insert a provision that explicitly allows for public interest considerations, rather than just cost recovery, to be applied when fees are set. The Treasury Board included public interest considerations in its guidelines for the User Fees Act. But as a fundamental principle, this should be written into the Act itself, rather than relying on subsequent regulations or guidelines.
  • Require that any proposed increases that are twice the annual consumer price index adjustment, and that directly impact the public (e.g., passport fees, park fees), be referred to the relevant Parliamentary committee for review in advance.
  • Ensure that Treasury Board Secretariat regulations and guidelines reflect these two points as good practices, even if the government chooses not to amend the Act.

While there may have been a need to streamline the consultation process for fee increases, the proposed “Service Fees Act” makes it too easy for government to raise fees without any meaningful public consultation and debate. The government needs to ensure a reasonable balance between efficiency and consultation, particularly for those fees that affect the general public.

Source: The pendulum swing on government service fees

House/Senate Brief: Bill C-44 Division 21: Risks and Implications of the Service Fees Act

Diversity in the Senate – My latest in Policy Options

My latest, analyzing the diversity of Senators. Intro teaser below:

With the large number of Senate appointments made by Prime Minister Justin Trudeau and a more independent role for individual senators, a look at the current level of diversity in the Upper Chamber is timely.

In essence, the Trudeau appointments have made the Senate more diverse in terms of gender, visible minorities, and Indigenous peoples and thus more representative of the people it serves.

However, when viewed from the perspective of education and occupation, there is less diversity: more senators with higher degrees, and more senators with an activist background and less with a business background.

Given the increased independence of senators, the increased ethnic and gender diversity and decreased educational and occupational diversity    may play a role in terms of how the Senate responds to legislation and plays its sober second thought role.

This analysis contrasts the various aspects of diversity between the 43 non-affiliated senators (33 form the Independent Senators Group, of whom 28 were appointed by Prime Minister Trudeau),  and the 41 Conservative and 21 Liberal senators (December 2016).

Source: Diversity in the Senate – Policy Options

Will changes to the Citizenship Act affect the naturalization rate? – Policy Options

My piece in IRPP’s Perspectives on the need for the Senate to consider the decline in the naturalization rate in its review of Bill C-6’s changes to the Citizenship Act.

Source: Will changes to the Citizenship Act affect naturalization rate? – Policy Options

Integration and multiculturalism: Finding a new metaphor – Policy Options

Having never been comfortable with the “two-way street” metaphor for immigrant integration, I finally got around to articulating my reasons more formally, and proposing some alternatives and stimulating debate and discussion.

The following article in IRPP’s Policy Options is the result.

Some other alternatives that I have received to date include: Wittgensteinian sailing a boat while always renewing the planks, in economics, the supply and demand model in setting prices and achieving a new equilibrium, big river with little streams, and the Norwegian metaphor of samspill, or play together.

I welcome additional suggestions and comments:

When immigration officials and advocates talk about the integration process for new Canadians, often they reach for the metaphor of a “two-way street.” Immigration, Refugees and Citizenship Canada notes that “the focus of the integration programs at CIC is on operationalizing the ‘two-way street approach’…through assisting individuals to become active, connected and productive citizens.” The Canadian Index on Immigration Integration talks about the “metaphorical meeting of the immigrant and the receiving population somewhere in the middle of the street.” Former Immigration minister Jason Kenney used the phrase in speeches, where he emphasized the duty of newcomers to adapt to their new home and the responsibility of Canadians to accommodate them.

But does a metaphor matter? It does, given Canada’s increasingly diverse population and the challenges of ensuring successful integration of immigrants and their children. Getting the metaphor right will convey expectations to newcomers and the host society alike regarding their respective roles in the integration process.

The “two-way street” is handy shorthand. With integration, immigrants and the host society both adapt to each other’s presence, instead of the assimilationist expectation that immigrants will completely adopt the values, attitudes, norms and lifestyles of the host. The aim of integration is for immigrants and their children to obey Canadian laws and eventually and voluntarily to adopt Canadian values, attitudes, norms and lifestyles, most notably gender equality and acceptance of other groups. The ultimate goal is that they eventually enjoy economic, social, health and political outcomes that are broadly comparable with those of “old-stock” Canadians. Accommodation by the host society aims at facilitating that integration process.

While diversions from this norm exist; for example, voluntary exclusion as in the cases of Mennonites or other religious communities, or involuntary exclusion because of racism and discrimination (separation or segregation, to use psychology scholar John Berry’s terms), the overall emphasis is on this integration process.

However, in order for metaphors to be successful, they have to speak to the broader population, not just the specialists, and they have to resonate in a visceral way. They have to be simple and easy to grasp, but they also have to avoid being simplistic. The “two-way-street” metaphor is unsuccessful for a number of reasons.

Streets are for cars, not people. Cars do not meet on a two-way street, they drive past each other (and if they do meet, usually it is in a crash!). Integration is about mingling, interacting and adapting.

The integration process is asymmetric: it is more important for immigrants and new Canadians to adapt to Canadian laws, norms and values than it is for the host society to adjust to them. The meeting point is not “somewhere in the middle” between the host society and the newcomers, but much closer to the host society (80/20 percent, in my view).

And integration is dynamic and ever changing, not linear. What is deemed acceptable evolves through the democratic, legal and political processes, as newcomers assert their identities and their rights and the host society responds. What was originally considered unacceptable (e.g., turban-wearing Mounties) becomes normal; what might initially be considered acceptable (e.g., sharia family law in Ontario) is deemed unacceptable following public discussion and debate.

It also must be noted that the values of the host society are dynamic and change over time; for example, attitudes toward gender equality, homosexuality and same sex marriage have undergone a sea change in Canada and elsewhere over the past generation. Immigrants and new Canadians are expected eventually to follow suit.

The metaphor also fails to capture the diversity and multilayered identities in Canadians, and how these can change with contexts and individual preferences. Ethnicity, religion, gender, age, generation, and lived experiences all play a part at the individual and group levels.

So, what are better phrases that describe this process and dynamic, and how do these capture the roles of newcomers and the host society?

Similar to the multiculturalism dynamic, the integration/accommodation dynamic captures the fact that newcomers integrate into the host society while the host society accommodates and adjusts to newcomer needs and identities. A combination of push and pull between newcomers and the host society, mediated through political, judicial and everyday socio-cultural processes, finds an equilibrium point, which evolves over time. Individuals and groups provide the push that forces the host society to respond. And, of course, as the host society itself becomes more diverse and its values evolve, this mediation happens between not only the “old stock” Canadians of the host society but also the more established and more recent groups of immigrants.

A more sophisticated version of this characterization can be seen in Harald Bauder’s description of integration as a process of Hegelian dialectic: thesis, antithesis, synthesis, or the host society (thesis) being challenged by newcomers (antithesis), resulting in a new balance and accommodation (synthesis), hence reducing the difference between the “newcomer ‘other’ and the Canadian ‘self’.”

Another concept used by some academics and policy-makers in a multiculturalism context, is that of harmony/jazz. Harmony is provided by the legal and constitutional framework that applies to all Canadians — it is the underlying melody. Jazz reflects improvisation in dealing with accommodation requests within that overall context. Again, the underlying harmony predominates. This approach has the advantage of being more flexible in dealing with accommodation pressures.

So what best describes integration and multiculturalism? Passing each other on a two-way street or “making music together?” My preference is the latter, as it describes the dynamic, complex and varied nature of how we, in our ever more diverse society, continue to search for a new and hopefully improved equilibrium.  Perhaps just as every jazz band interprets a musical composition in new and unanticipated ways, so does a healthy, diverse multicultural society.

I would like to thank Michael Adams and Harald Bauder for their helpful comments and suggestions on integration definitions and metaphors.

Source: Integration and multiculturalism: Finding a new metaphor – Policy Options