Connect with us

Techcrunch

Tips for applying an intersectional framework to AI development – TechCrunch

Ultimately, there isn’t a “one-size-fits-all” approach to building intersectional AI. At The Trevor Project, our team has outlined a methodology based on what we do, what we know today and the specific communities we serve.

Published

on

Kendra Gaunt (she/her or they/them pronouns) is a data and AI product owner at The Trevor Project, the world’s largest suicide prevention and crisis intervention organization for LGBTQ youth. A 2019 Google AI Impact Grantee, the organization is implementing new AI applications to scale its impact and save more young LGBTQ lives.

By now, most of us in tech know that the inherent bias we possess as humans creates an inherent bias in AI applications — applications that have become so sophisticated they’re able to shape the nature of our everyday lives and even influence our decision-making.

The more prevalent and powerful AI systems become, the sooner the industry must address questions like: What can we do to move away from using AI/ML models that demonstrate unfair bias?

How can we apply an intersectional framework to build AI for all people, knowing that different individuals are affected by and interact with AI in different ways based on the converging identities they hold?

Start with identifying the variety of voices that will interact with your model. Intersectionality: What it means and why it matters

Before tackling the tough questions, it’s important to take a step back and define “intersectionality.” A term defined by Kimberlé Crenshaw, it’s a framework that empowers us to consider how someone’s distinct identities come together and shape the ways in which they experience and are perceived in the world.

This includes the resulting biases and privileges that are associated with each distinct identity. Many of us may hold more than one marginalized identity and, as a result, we’re familiar with the compounding effect that occurs when these identities are layered on top of one another.

At The Trevor Project, the world’s largest suicide prevention and crisis intervention organization for LGBTQ youth, our chief mission is to provide support to each and every LGBTQ young person who needs it, and we know that those who are transgender and nonbinary and/or Black, Indigenous, and people of color face unique stressors and challenges.

So, when our tech team set out to develop AI to serve and exist within this diverse community — namely to better assess suicide risk and deliver a consistently high quality of care — we had to be conscious of avoiding outcomes that would reinforce existing barriers to mental health resources like a lack of cultural competency or unfair biases like assuming someone’s gender based on the contact information presented.

Though our organization serves a particularly diverse population, underlying biases can exist in any context and negatively impact any group of people. As a result, all tech teams can and should aspire to build fair, intersectional AI models, because intersectionality is the key to fostering inclusive communities and building tools that serve people from all backgrounds more effectively.

Doing so starts with identifying the variety of voices that will interact with your model, in addition to the groups for which these various identities overlap. Defining the opportunity you’re solving is the first step because once you understand who is impacted by the problem, you can identify a solution. Next, map the end-to-end experience journey to learn the points where these people interact with the model. From there, there are strategies every organization, startup and enterprise can apply to weave intersectionality into every phase of AI development — from training to evaluation to feedback.

Datasets and training

The quality of a model’s output relies on the data on which it’s trained. Datasets can contain inherent bias due to the nature of their collection, measurement and annotation — all of which are rooted in human decision-making. For example, a 2019 study found that a healthcare risk-prediction algorithm demonstrated racial bias because it relied on a faulty dataset for determining need. As a result, eligible Black patients received lower risk scores in comparison to white patients, ultimately making them less likely to be selected for high-risk care management.

Fair systems are built by training a model on datasets that reflect the people who will be interacting with the model. It also means recognizing where there are gaps in your data for people who may be underserved. However, there’s a larger conversation to be had about the overall lack of data representing marginalized people — it’s a systemic problem that must be addressed as such, because sparsity of data can obscure both whether systems are fair and whether the needs of underrepresented groups are being met.

To start analyzing this for your organization, consider the size and source of your data to identify what biases, skews or mistakes are built-in and how the data can be improved going forward.

The problem of bias in datasets can also be addressed by amplifying or boosting specific intersectional data inputs, as your organization defines it. Doing this early on will inform your model’s training formula and help your system stay as objective as possible — otherwise, your training formula may be unintentionally optimized to produce irrelevant results.

At The Trevor Project, we may need to amplify signals from demographics that we know disproportionately find it hard to access mental health services, or for demographics that have small sample sizes of data compared to other groups. Without this crucial step, our model could produce outcomes irrelevant to our users.

Evaluation

Model evaluation is an ongoing process that helps organizations respond to ever-changing environments. Evaluating fairness began with looking at a single dimension — like race or gender or ethnicity. The next step for the tech industry is figuring out how to best compare intersectional groupings to evaluate fairness across all identities.

To measure fairness, try defining intersectional groups that could be at a disadvantage and the ones that may have an advantage, and then examine whether certain metrics (for example, false-negative rates) vary among them. What do these inconsistencies tell you? How else can you further examine which groups are underrepresented in a system and why? These are the kinds of questions to ask at this phase of development.

Developing and monitoring a model based on the demographics it serves from the start is the best way for organizations to achieve fairness and alleviate unfair bias. Based on the evaluation outcome, a next step might be to purposefully overserve statistically underrepresented groups to facilitate training a model that minimizes unfair bias. Since algorithms can lack impartiality due to societal conditions, designing for fairness from the outset helps ensure equal treatment of all groups of individuals.

Feedback and collaboration

Teams should also have a diverse group of people involved in developing and reviewing AI products — people who are diverse not only in identities, but also in skillset, exposure to the product, years of experience and more. Consult stakeholders and those who are impacted by the system for identifying problems and biases.

Lean on engineers when brainstorming solutions. For defining intersectional groupings, at The Trevor Project, we worked across the teams closest to our crisis-intervention programs and the people using them — like Research, Crisis Services and Technology. And reach back out to stakeholders and people interacting with the system to collect feedback upon launch.

Ultimately, there isn’t a “one-size-fits-all” approach to building intersectional AI. At The Trevor Project, our team has outlined a methodology based on what we do, what we know today and the specific communities we serve. This is not a static approach and we remain open to evolving as we learn more. While other organizations may take a different approach to build intersectional AI, we all have a moral responsibility to construct fairer AI systems, because AI has the power to highlight — and worse, magnify — the unfair biases that exist in society.

Depending on the use case and community in which an AI system exists, the magnification of certain biases can result in detrimental outcomes for groups of people who may already face marginalization. At the same time, AI also has the ability to improve quality of life for all people when developed through an intersectional framework. At The Trevor Project, we strongly encourage tech teams, domain experts and decision-makers to think deeply about codifying a set of guiding principles to initiate industry-wide change — and to ensure future AI models reflect the communities they serve.

Source: https://techcrunch.com/2020/12/18/tips-for-applying-an-intersectional-framework-to-ai-development/

tips-for-applying-an-intersectional-framework-to-ai-development-–-techcrunch

Techcrunch

Golden Gate Ventures forecasts a record number of exits in Southeast Asia – TechCrunch

Despite the pandemic’s economic impact, Southeast Asia’s startup ecosystem has proven to be very resilient. In fact, a new report from investment firm Golden Gate Ventures predicts a record number of exits will happen in the region over the next couple of years, thanks to factors like a maturing ecosystem, more secondary buyers and the […]

Published

on

Despite the pandemic’s economic impact, Southeast Asia’s startup ecosystem has proven to be very resilient. In fact, a new report from investment firm Golden Gate Ventures predicts a record number of exits will happen in the region over the next couple of years, thanks to factors like a maturing ecosystem, more secondary buyers and the emergence of SPACs.

The firm’s comprehensive “Southeast Asia Exit Landscape Report 2.0,” is a followup to a previous report published in 2019.

Here are some highlights from the latest report, along with additional insight from Golden Gate Ventures partner Michael Lints, its lead author. For both reports, Golden Gate Ventures partnered with business school INSEAD to survey general and limited partners in the region. It also draws on Golden Gate Ventures’ proprietary database, which dates back to 2012 and tracks information like the time between funding rounds and fundraising success rates, as well as public databases, reports and expert commentary from the New York Stock Exchange.

The overall exit landscape

Despite the pandemic’s economic impact, tech proved to be resilient globally (for example, there were a number of initial public offers in the United States at record prices). While Southeast Asia’s tech ecosystem is relatively younger, Lints told TechCrunch its resiliency was driven by companies founded years ago that suddenly saw an increase in demand for their services because of the pandemic.

“We’ve built infrastructure over the past eight to nine years, when it comes to e-commerce, logistics, some on the healthcare side as well, and when the pandemic happened, people were suddenly stuck at home,” Lints said. He added “If you look at the pickup for most of the e-commerce companies, they at least doubled their revenue. For last-mile logistics companies, they’ve increased their revenue. There was a lot of pickup on the digital healthcare side as well.”

While tech fared well compare to many other industries, one downside was that the COVID-19 pandemic caused overall global venture capital investment to decline. Southeast Asia’s startup ecosystem was not immune, and had less exits, but it still did relatively well, with $8.2 billion invested in 2020, according to a report by Cento Ventures and Tech In Asia.

It’s important to note that more than half of that funding was raised in very large rounds by unicorns like Grab, Go-jek and Traveloka, but Cento Ventures found there was also an increase in investments between $50 million to $100 million for other startups. These are usually Series B and C rounds, which Golden Gate Ventures says creates a strong pipeline for potential exits over the next three to four years.

“If you go back even just two years, the amount of B rounds that are happening now, I’ve never seen that number before. It’s a definite increase,” said Lints.

Investments are also continuing to flow into Southeast Asia. According to the report, there was $6 billion of funding in just the first quarter of 2021 (based on data from DealStreet Asia, PWC and Genesis Ventures), making it the strongest start to a year in the region’s history.

This bodes well for the possibility of mergers and acquisitions in 2021. The report found that there were less exits in 2019 and 2020 than in 2018, but not just because of the pandemic—many startups wanted to remain venture-backed for longer. Golden Gate Ventures expects M&A activity will pick up again. In 2021, it forecasts acquisition deals worth more than $30 million, large mergers and an increase in SPACs.

What’s in the pipeline

Golden Gate Ventures predicts that a total of 468 startup exits will happen between 2020 and 2022, compared to the 412 forecast in the previous edition of its report. This is due to more late-stage private equity investors, including secondary buyers, SPACs and a welcoming public market.

Lints said secondary buyers will include a mix of family offices, conglomerates and venture funds that want a higher allocation in a company or to pre-empt a forthcoming round.

“What I think is interesting is some of the later-stage funds, so private equity funds, and not only ones that are in Southeast Asia, but even foreign ones, are now looking to get a position in companies that they assume will be able to raise a Series D or Series E over the next few years. That’s something I haven’t seen before, it’s relatively new in the market,” he added.

Golden Gate Ventures expects M&A activity to continue being the main way Southeast Asian startups exit, potentially accounting for up to 80% of deals, followed by secondary sales (15%) and IPOs (5%).

In fact, there was a record number of M&A deals in 2020, despite the pandemic. Golden Gate Ventures estimates that 45 deals happened, especially in e-commerce, fintech, media, adtech and social networking, as larger companies acquired startups to grow their tech stacks.

More companies going public will create a cascading effect through Southeast Asia’s ecosystem. The report forecasts that companies like Gojek and Trax, who have already made several high-profile acquisitions, will continue buying startups if they list publicly and have more liquidity.

Series B and C deals

While there will be more exits, there are also more opportunities for companies to raise larger later-stage rounds to stay private, if they want to—a sign of Southeast Asia’s maturing ecosystem, said Lints.

As the pandemic unfolded in 2020, the number of pre-seed and seed deals fell. On the other hand, the report found that it became quicker for startups to raise Series B or C rounds, or less than 21 months on average.

“If you look at typical exits between 2015 to 2017, you could argue that some of those exits might have been too early because the company was still in a growth trajectory, but there was hardly any follow-on funding for them to expand to a new country, for instance, or build out a new product,” said Lints. “So their only revenue to raise money was to be acquired by a larger company so they could keep building the product.”

“I think now you’re able to raise that Series C round, which allows you to expand the company and stay private, as opposed to having to drive towards an exit,” he added. “I think that shows the maturity of the ecosystem now and, again, it’s a huge advantage because founders have these amazing things they want to build, and now actually have the capital to do so and to really try to compete, and that has definitely been a big change.”

Another good thing is that the increase in later-stage funding does not appear to be creating a pre-seed and seed funding gap. This is partly because early employees from mature companies that have raised massive rounds often branch out and become founders themselves. As they launch startups, they have the benefit of being familiar with how fundraising works and a network. For example, a significant number of alumni from Grab, Gojek and Lazada have gone on to found companies.

“They seem to be raising a lot faster, and I think the second thing that’s happening across the board is we’re seeing more scouts putting really early checks into companies,” said Lints. “My assumption is if you look at the Series A pipeline, which is still pretty long, that has to come from a large number of pre-seed and seed deals.”

Funds want to cash out

Another factor that may drive an increase in exits—especially M&A deals—are funds that have reached the point where they want to cash out. Golden Gate Ventures’ 2019 report forecast that the first batch of institutional venture funds launched in 2010 to 2012 will start reaching the end of their lifecycle in 2020. This means the general partners of these funds are exploring exit opportunities for their portfolios, leading to an increase in secondary and M&A deals.

This in turn will increase the number of secondary markets, which have typically been low in Southeast Asia. The original investors won’t necessarily push for portfolio companies to sell themselves, but instead look at secondary buyers who might be keen on mergers and M&A deals.

“The thing we’ve seen over the last 18 months is there’s been a larger pickup in the secondary markets, where later-stage investors, in some cases family-owned businesses or family offices, are looking to get access to deals that were started eight, nine or 10 years ago. You’ll see the cap tables of these companies change, and that does mean the founders will have different shareholders,” said Lints.

“These are typically for companies that are performing well, where you can foresee that they will be able to fundraise within the next 12 months. For the ones that are in a more difficult position, I think it’s going to be tricky,” he added. “When you have a portfolio of companies as a fund, that doesn’t necessarily mean that you can sell all 20 of them, so I think for some founders, the impact will be that they will need to make a decision to continue the business and buy back the shares their investors are holding, or are they going to liquidate the business or look for a trade sale.”

SPAC opportunities

The biggest SPAC news in Southeast Asia was Grab’s announcement it will go public in the United States following a $40 billion SPAC deal. Lints expects more Southeast Asian companies to take the SPAC route when going public. Not only does the process give them more flexibility, but for startups that want to list in the U.S., working with a SPAC can help them.

“My guess is with New York allowing direct listings, I think more and more people will shy away from the traditional IPO route and look at what is the fastest and most flexible way to list on a stock exchange. For Southeast Asia, listing has never been easy, so I think SPACs will definitely open the floodgates,” said Lints.

Barriers not only include regulatory filings, pre-IPO roadshows and high costs, but also “concern whether the international retail investor or public markets actually understand these companies in Southeast Asia,” he added. “If you have a very strong sponsor team that is running the SPAC, they can be super helpful in positioning the company, doing the marketing and getting interest from the market as well.”

Both the Singapore Exchange and Indonesian Stock Exchange are preparing to allow SPACs in an effort to attract more tech listings.

Lints said this will allow companies to consider a dual listing in Southeast Asia and the U.S. for larger returns. “A dual listing would be an amazing option and I think through the avenue of SPACs, that makes a lot of sense.”

While tech fared well compare to many other industries, one downside was that the COVID-19 pandemic caused overall global venture capital investment to decline. Southeast Asia’s startup ecosystem was not immune, and had less exits, but it still did relatively well, with $8.2 billion invested in 2020, according to a report by Cento Ventures and Tech In Asia.

Source: https://techcrunch.com/2021/06/15/golden-gate-ventures-forecasts-a-record-number-of-exits-in-southeast-asia/

golden-gate-ventures-forecasts-a-record-number-of-exits-in-southeast-asia-–-techcrunch

Continue Reading

Techcrunch

Jeff Bezos’ Blue Origin auctions off seat on first human spaceflight for $28M – TechCrunch

Blue Origin has its winning bidder for its first ever human spaceflight, and the winner will pay $28 million for the privilege of flying aboard the company’s debut private astronaut mission. The winning bid came in today during a live auction, which saw 7,600 registered bidders, from 159 countries compete for the spot. This was […]

Published

on

Blue Origin has its winning bidder for its first ever human spaceflight, and the winner will pay $28 million for the privilege of flying aboard the company’s debut private astronaut mission. The winning bid came in today during a live auction, which saw 7,600 registered bidders, from 159 countries compete for the spot.

This was the culmination of Blue Origin’s three part bidding process for the ticket, which included a blind auction first, followed by an open, asynchronous auction with the highest bid posted to the company’s website whenever it changed. This last live auction greatly ramped up the value of the winning bid, which was at just under $5 million prior to the event.

This first seat up for sale went for a lot more than what an actual, commercial spot is likely to cost on Blue Origin’s New Shepard capsule, which flies to suborbital space and only spends a few minutes there before returning to Earth. Estimates put the cost of a typical launch at someone under $1 million, likely closer to $500,000 or so. But this is the first, which is obviously a special distinction, and it’s also a trip that will allow the winning bidder to pretty much literally rub elbows with Blue Origin founder Jeff Bezos, who is going to be on the flight as well, along with his brother Mark, and a fourth passenger that Blue Origin says it will be announcing sometime in the coming “weeks,” ahead of the July 20 target flight date.

As for who won the auction, we’ll also have to wait to find that out, since the winner’s identity is also going to be “released in the weeks following” the end of today’s live bidding. And in case you thought that $28 million might represent a big revenue windfall for Blue Origin, which has spent years developing its human spaceflight capability, think again: The company is donating it to its Club for the Future non-profit foundation, which is focused on encouraging kids to pursue careers in STEM in a long-term bid to help Bezos’ larger goals of making humanity a spacefaring civilization.

You can re-watch the entire live bidding portion of the auction via the stream below.

As for who won the auction, we’ll also have to wait to find that out, since the winner’s identity is also going to be “released in the weeks following” the end of today’s live bidding. And in case you thought that $28 million might represent a big revenue windfall for Blue Origin, which has spent years developing its human spaceflight capability, think again: The company is donating it to its Club for the Future non-profit foundation, which is focused on encouraging kids to pursue careers in STEM in a long-term bid to help Bezos’ larger goals of making humanity a spacefaring civilization.

Source: https://techcrunch.com/2021/06/12/jeff-bezos-blue-origin-auctions-off-seat-on-first-human-spaceflight-for-28m/

jeff-bezos’-blue-origin-auctions-off-seat-on-first-human-spaceflight-for-$28m-–-techcrunch

Continue Reading

Techcrunch

UBS investment makes Byju’s the most valuable startup in India – TechCrunch

Edtech giant Byju’s has become the most valuable startup in India after raising about $350 million in a new tranche of investment from UBS Group and Zoom founder Eric Yuan, Blackstone and others that valued the Bangalore-based firm at $16.5 billion (post-money). In a new filing, Byju’s revealed that scores of investors including Abu Dhabi […]

Published

on

Edtech giant Byju’s has become the most valuable startup in India after raising about $350 million in a new tranche of investment from UBS Group and Zoom founder Eric Yuan, Blackstone and others that valued the Bangalore-based firm at $16.5 billion (post-money).

In a new filing, Byju’s revealed that scores of investors including Abu Dhabi government fund ADQ and Phoenix Rising had together invested about $350 million in the startup. The new valuation helps Byju’s surpass Paytm, which was last valued at $16 billion, for the crown position in the Indian startup ecosystem. (Paytm is currently working on exploring the public markets and eyeing to raise as much as $3 billion and eyeing a valuation of up to $30 billion.)

The new tranche of investment is part of a larger round that Byju’s kickstarted earlier this year and is looking to secure over $1.5 billion. Some of its recent investors also include B Capital Group and hedge fund XN. The startup was valued at $11 billion late last year, and $5.75 billion in July 2019.

The startup plans to use the fresh capital, in part, to acquire more startups. Byju’s, which acquired Indian physical coaching institute Aakash for nearly $1 billion earlier this year, is conducting due diligence to buy and online learning startup Toppr and has also engaged with U.S.-based Epic, TechCrunch reported earlier this year.

Byju’s prepares students pursuing undergraduate and graduate-level courses, and in recent years it has also expanded its catalog to serve all school-going students. Tutors on the Byju’s app tackle complex subjects using real-life objects such as pizza and cake.

The pandemic, which prompted New Delhi to enforce a months-long nationwide lockdown and close schools, accelerated its growth, and those of several other online learning startups including Unacademy and Vedantu.

As of early this year, Byju’s said it had amassed over 80 million users, 5.5 million of whom are paying subscribers. Byju’s, which is profitable, generated revenue of over $100 million in the U.S. last year, Deborah Quazzo, managing partner of GSV Ventures (which has backed the Indian startup), said at a session in March held by Indian venture fund Blume Ventures.

The startup executives said at a UBS event earlier this year that Byju’s current revenue run rate is $800 million, a figure they expect will reach $1 billion in the next 12-15 months. It has also accelerated its international expansion plans in recent months.

Source: https://techcrunch.com/2021/06/12/ubs-investment-makes-byjus-the-most-valuable-startup-in-india/

ubs-investment-makes-byju’s-the-most-valuable-startup-in-india-–-techcrunch

Continue Reading

Title

CNBC3 hours ago

Oracle guidance misses expectations, stock drops

Oracle reported better-than-expected results and showed accelerating growth compared with the immediate impact of the coronavirus last year.

Ventureburn16 hours ago

Wayja releases SA’s first peer-to-peer betting app

Wayja launches its cashless peer-to-peer betting app, available on the Wayja site and on all major app stores by December...

ZDNET22 hours ago

Apple releases emergency update for older iPhones and iPads

If you're running iOS 12, this is an update for you.

Crunchbase1 day ago

Macrometa Locks Down $20M To Be The Amazon Prime Of Edge Computing

Palo Alto, California-based edge compute company Macrometa closed a $20 million Series A less than eight months after announcing its...

Cointelegraph1 day ago

Crypto miners eye cheap power in Texas, but fears aired over impact on the grid

Can Texas meet the electricity demands of migrating Chinese Bitcoin miners?

Coinpedia2 days ago

Bitcoin Cash Price Prediction, Will BCH Hit Incredible Surges At $1000?

According to Coinpedia's formulated Bitcoin Cash price prediction, the coin's price may strike a maximum of $1417.33 by the year...

Blockchain news2 days ago

US Space Force Makes its Foray into the NFT Metaverse

The United States Space Force is launching an NFT series named after Neil Armstrong.

Reuters2 days ago

EXCLUSIVE Galp to hold off on LNG investment until Mozambique ensures security

Portugal's Galp Energia (GALP.LS), a partner in an Exxon Mobil-led gas consortium in Mozambique, will not invest in onshore plants...

Techcrunch2 days ago

Golden Gate Ventures forecasts a record number of exits in Southeast Asia – TechCrunch

Despite the pandemic’s economic impact, Southeast Asia’s startup ecosystem has proven to be very resilient. In fact, a new report...

Bioengineer2 days ago

Physical activity reduces cardiovascular risk in rheumatic patients

People with diseases such as rheumatoid arthritis and lupus are more likely to have heart attacks, angina, and strokes. A

Review

    Select language

    Trending