Smoke is rising from Apple’s Conclave

by Cherian Abraham 9 min read October 2, 2013

TL;DRRead within as to how Touch ID is made possible via ARM’s TrustZone/TEE, and why this matters in the context of the coming Apple’s identity framework. Also I explain why primary/co-processor combos are here to stay. I believe that eventually, Touch ID has a payments angle – but focusing on e-commerce before retail. Carriers will weep over a lost opportunity while through Touch ID, we have front row seats to Apple’s enterprise strategy, its payment strategy and beyond all – the future direction of its computing platform.

I hadshared my takeon a possible Apple Biometric solution during the Jan of this year based on its Authentec acquisition. I came pretty close, except for the suggestion that NFC is likely to be included. (Sigh.)

Its a bit early to play fast and loose with Apple predictions, but its Authentec acquisition should rear its head sometime in the near future (2013 – considering Apple’s manufacturing lead times), that a biometric solution packaged neatly with an NFC chip and secure element could address three factors that has held back customer adoption of biometrics: Ubiquity of readers, Issues around secure local storage and retrieval of biometric data, Standardization in accessing and communicating said data. An on-chip secure solution to store biometric data – in the phone’s secure element can address qualms around a central database of biometric data open to all sorts of malicious attacks. Standard methods to store and retrieve credentials stored in the SE will apply here as well.

Why didn’t Apple open up Touch ID to third party dev?

Apple expects a short bumpy climb ahead for Touch ID before it stabilizes, as early users begin to use it. By keeping its use limited to authenticating to the device, and to iTunes – it can tightly control the potential issues as they arise. If Touch ID launched with third party apps and were buggy, it’s likely that customers will be confused where to report issues and who to blame.

That’s not to say that it won’t open up Touch ID outside of Apple. I believe it will provide fettered access based on the type of app and the type of action that follows user authentication. Banking, Payment, Productivity, Social sharing and Shopping apps should come first. Your fart apps? Probably never.

Apple could also allow users to set their preferences (for app categories, based on user’s current location etc.) such that biometrics is how one authenticates for transactions with risk vs not requiring it. If you are at home and buying an app for a buck – don’t ask to authenticate. But if you were initiating a money transfer – then you would. Even better – pair biometrics with your pin for better security.Chip and Pin? So passé.

Digital Signatures, iPads and the DRM 2.0:

It won’t be long before an iPad shows up in the wild sporting Touch ID. And with Blackberry’s much awaited and celebrated demise in the enterprise, Apple will be waiting on the sidelines – now with capabilities that allowdigital signatures to become ubiquitous and simple – on email, contracts or anything worth putting a signature on. Apple has already made its iWork productivity apps(Pages, Numbers, Keynote), iMovie and iPhotofreefor new iOS devices activated w/ iOS7.

Apple, with a core fan base that includes photographers, designers and other creative types, can now further enable iPads and iPhones to become content creation devices, with the ability to attribute any digital content back to its creator by a set of biometric keys. Imagine a new way to digitally create and sign content, to freely share, without worrying about attribution.

Further Apple’s existing DRM frameworks are strengthened with the ability to tag digital content that you download with your own set of biometric keys. Forget disallowing sharing content –Apple now has a way to create a secondary marketplace for its customers to resell or loan digital content, and drive incremental revenue for itself and content owners.

Conclaves blowing smoke:

In a day and age where we forego the device for storing credentials – whether it be due to convenience or ease of implementation – Apple opted for an on-device answer for where to store user’s biometric keys. There is a reason why it opted to do so – other than the obvious brouhaha that would have resulted if it chose to store these keys on the cloud.

Keys inside the device. Signed content on the cloud. Best of both worlds.

Biometric keys need to be held locally, so that authentication requires no roundtrip and therefore imposes no latency. Apple would have chosen local storage (ARM’s SecurCore) as a matter of customer experience, and what would happen if the customer was out-of-pocket with no internet access. There is also the obvious question that a centralized biometric keystore will be on the crosshairs of every malicious entity. By decentralizing it, Apple made it infinitely more difficult to scale an attack or potential vulnerability.

More than the A7, the trojan in Apple’s announcement was the M7 chip – referred to as the motion co-processor. I believe the M7 chip does more than just measuring motion data.

M7 – A security co-processor?

I am positing that Apple is using ARM’sTrustZonefoundation and it may be using the A7 or the new M7 co-processor for storing these keys and handling the secure backend processing required.

Horace Dediu of Asymco hadcalled to questionwhy Apple had opted for M7 and suggested that it may have a yet un-stated use.I believe M7 is not just a motion co-processor, it is also a security co-processor. I am guessing M7 is based on the Cortex-M series processors and offloads much of this secure backend logic from the primary A7 processor and itmay bethat the keys themselves are likely to be stored here on M7. The Cortex-M4 chip has capabilities that sound very similar to what Apple announced around M7 – such as very low power chip, that is built to integrate sensor output and wake up only when something interesting happens. We should know soon.

This type of combo – splitting functions to be offloaded to different cores, allows each cores to focus on the function that it’s supposed to performed. I suspect Android will not be far behind in its adoption, where each core focuses on one or more specific layers of the Android software stack. Back at Google I/O 2013, it had announced 3 new APIs (the Fused location provider) that enables location tracking without the traditional heavy battery consumption. Looks to me that Android decoupled it so that we will see processor cores that focus on these functions specifically – soon.

ARMv8

I am fairly confident that Apple has opted for ARM’s Trustzone/TEE. Implementation details of the Trustzone are proprietary and therefore not public. Apple could have made revisions to the A7 chip spec and could have co-opted its own. But using the Trustzone/TEE and SecurCore allows Apple to adopt existing standards around accessing and communicating biometric data. Apple is fully aware of the need to mature iOS as a trusted enterprise computing platform – to address the lack of low-end x86 devices that has a hardware security platform tech. And this is a significant step towards that future.

What does Touch ID mean to Payments?

Apple plans for Touch ID kicks off with iTunes purchase authorizations. Beyond that, as iTunes continue to grow in to a media store behemoth – Touch ID has the potential to drive fraud risk down for Apple – and to further allow it to drive down risk as it batches up payment transactions to reduce interchange exposure. It’s quite likely that à la Walmart, Apple has negotiated rate reductions – but now they can assume more risk on the front-end because they are able to vouch for the authenticity of these transactions. As they say – customer can longer claim the fifth on those late-night weekend drunken purchase binges.

Along with payment aggregation, or via iTunes gift cards – Apple has now another mechanism to reduce its interchange and risk exposure. Now – imagine if Apple were to extend this capability beyond iTunes purchases – and allow app developers to process in-app purchases of physical goods or real-world experiences through iTunes in return for better blended rates? (instead of Paypal’s 4% + $0.30).Heck, Apple can opt for short-term lending if they are able to effectively answer the question of identity – as they can with Touch ID. It’s Paypal’s ‘Bill Me Later’ on steroids.

Effectively, a company like Apple who has seriously toyed with the idea of a Software-SIM and a “real-time wireless provider marketplace” where carriers bid against each other to provide you voice, messaging and data access for the day – and your phone picks the most optimal carrier, how far is that notion from picking the cheapest rate across networks for funneling your payment transactions? Based on the level of authentication provided or other known attributes – such as merchant type, location, fraud risk, customer payment history – iTunes can select across a variety of payment options to pick the one that is optimal for the app developer and for itself.

And finally, who had the most to lose with Apple’s Touch ID?

Carriers. Iwrote about this before as well, here’s what I wrote then (edited for brevity):

Does it mean that Carriers have no meaningful role to play in commerce? Au contraire. They do. But its around fraud and authentication. Its around Identity. … But they seem to be stuck imitating Google in figuring out a play at the front end of the purchase funnel, to become a consumer brand(Isis). The last thing they want to do is leave it to Apple to figure out the “Identity management” question, which the latter seems best equipped to answer by way of scale, the control it exerts in the ecosystem, its vertical integration strategy that allows it to fold in biometrics meaningfully in to its lineup, and to start with its own services to offer customer value.

So there had to have been much ‘weeping and moaning and gnashing of the teeth’ on the Carrier fronts with this launch. Carriers have been so focused on carving out a place in payments, that they lost track of what’s important – that once you have solved authentication, payments is nothing but accounting. I didn’t say that. Ross Anderson of Kansas City Fed did.

What about NFC?

I don’t have a bloody clue. Maybe iPhone6? iPhone

This is a re-post from Cherian'soriginal blog post"Smoke is rising from Apple's Conclave"

Related Posts

Ask the Expert: A Closer Look at Modern Lending with Jeff Hops and Erin Haselkorn

In this first episode of Ask the Expert, Experian's Jeff Hops, Senior Director of Data Platform and Product, and Erin Haselkorn, Senior Director of Analyst Relations, explore how broader data and new signals can help lenders better understand today’s consumers, while maintaining responsible decisioning. Lending is changing  Interest rates, regulation, embedded finance and AI are reshaping the lending landscape. Consumer behavior is evolving just as quickly. But the core job hasn’t changed. Lenders are still making decisions about people they don’t fully know, and that makes data more important than ever. "There are periods where nothing changes, and periods where it seems like everything changes. We’re in the latter … but the core premise hasn’t changed. You’re still trying to lend to somebody you don’t know."Jeff Hops, Senior Director of Data Platform and Product To make those decisions with confidence, lenders need a strong foundation of identity, history and reliable signals. In a period of rapid change, the quality and completeness of that data become even more critical. A more complex view of today’s consumer What has changed is the consumer. Traditional credit data is foundational but can be further enhanced with visibility on how people earn, manage and move money. Income may come from multiple sources, and financial activity often spans bank accounts, applications (apps) and digital channels. Cash flow data, for example, can provide a clearer view of what’s actually coming into a consumer’s account, beyond what traditional records may show.These additional signals can help lenders better understand: Income variability across multiple earning sources Current financial behavior through cash flow activity Digital and identity-linked activity across channels These signals don’t replace traditional data; they expand it. The result is a more complete and current view of the consumer. From exploration to real-world application The conversation around broader data signals has moved beyond theory. Lenders are no longer just asking whether these signals are useful. They’re asking where, how and under what governance they can be applied across the lending lifecycle. Lenders are actively researching, testing and implementing new data sources across the lending lifecycle. What was once experimental is now operational. Institutions are progressing through a clear path: Research Understanding available signals and use cases Testing Evaluating performance in controlled environments Implementation Applying insights in production Today, alternative data is being used in areas like analytics, channel scoring and decisioning, often within governed environments that allow for safe testing and validation. AI may accelerate this shift by helping institutions identify patterns at scale, but its value depends on the strength of the underlying data: quality, governance, context and clear business use cases. More signal, more responsibility As data availability expands, lenders have access to more granular insights than ever before. That creates opportunity, but also responsibility. The institutions that lead won’t be the ones that use the most data. They’ll be the ones that know which signals to use, how to validate them and how to apply them in ways that are fair, explainable and aligned to consumer outcomes. “Institutions can unlock more granular and powerful decisions, but they have to do it responsibly.”Erin Haselkorn, Senior Director, Analyst Relations The future of lending will be shaped not just by how much data is available, but by how thoughtfully it’s applied. Keeping the consumer at the center of decisioning is essential to building trust and long-term success. Explore alternative data with us A more complete understanding of today’s consumers starts with better data. We help lenders responsibly incorporate broader data signals and advanced analytics into decisioning strategies, enhancing visibility into today’s consumers while strengthening risk assessment and expanding access to credit. Let’s work together to build more confident, more responsible lending decisions. Learn more Contact us About our experts Jeff Hops Senior Director, Data Platform and Product, Experian Jeff Hops is a Senior Director in Experian’s Financial Services and Data business with over eight years of experience driving innovation in credit and data solutions. He has led product development for Experian’s Credit Report and played a key role in launching Ascend Identity Platform™, a leading identity resolution platform. Erin Haselkorn Senior Director, Analyst Relations, Experian Erin Haselkorn is responsible for analyst relations for Experian. She has developed an understanding of key marketing trends across a broad range of verticals. Her market research around data strategy, AI, fraud, identity and data management, paired with her broad Experian product knowledge, gives her a unique understanding of business automation and data trends. Erin is a frequent spokesperson and guest blogger.

Published: June 22, 2026 by Julie.JLee@experian.com
How Consumer Vehicle Choices Are Shaping Automotive Loan Trends

Conversations about rising auto loan balances and higher monthly payments has often centered around increasing vehicle prices and elevated interest rates; and while those factors have undoubtedly played a role, another important piece of the puzzle is the type of vehicles consumers are choosing to purchase. According to Experian’s Automotive Consumer Trends Report: Q1 2026, consumers are continuing to opt for SUVs over other vehicle types, a trend that may be contributing to higher average loan amounts and monthly payments. SUVs accounted for 63.5% of all new retail vehicle registrations over the last 12 months, up from 62.8% a year ago. Additionally, more than 117 million SUVs were in operation across the United States in the first quarter of 2026, making up 42.2% of the market share. At the same time, traditional passenger cars continue to fall in share, coming in at 16.5%, a decrease from 18.4% last year. As consumers increasingly gravitate towards the larger vehicle segment, it reflects the ongoing desire for versatility, cargo capacity, and family-friendly functionality. Electrification’s growing role in consumer purchasing behavior Interestingly, electrified SUVs continue to gain traction, representing 27.7% of all new SUV registrations, these vehicles include battery-electric, hybrids, plug-in hybrids, and other alternative fuel types. Diving a bit deeper, the Tesla Model Y was the market share leader for new, retail electrified SUV registrations in the last 12 months, coming in at 15.8%. Rounding out the top five were Honda CR-V (9.6%), Toyota RAV4 (7.2%), Chevrolet Trax (7.2%), and Toyota Grand Highlander (3.4%). As model availability and familiarity with the electrification segment grows, the broader adoption of these vehicles are playing an increasingly important role in vehicle pricing and overall consumer demand. While average loan amounts and monthly payments are being driven by a combination of factors such as financing costs and consumer purchasing behavior, data in Q1 2026 demonstrates the continued interest in SUVs. This suggests that the industry’s shift toward larger vehicles is likely playing a meaningful role in today’s financing environment. To learn more about SUV insights, view the full Automotive Consumer Trends Report: Q1 2026 presentation.

Published: June 17, 2026 by Kirsten Von Busch
When New Data Impacts MBS Pricing: Student Loan Debt

In our previous post, we described the Current Second Lien Balance field, which is one of over 2,000 fields in the new Experian Mortgage Loan Performance (MLP) dataset. We showed that the Current Second Lien Balance field meets our three-pronged materiality standard for new data delivery: New: Provides information not available in existing datasets (i.e., orthogonal to currently available data). Material: Impacts a sizeable portion of the MBS universe. Significant: Differentiates collateral performance by a large enough margin to influence trading and risk management decisions. In this article, we discuss another field that satisfies the above criteria: Student Loan Balance.  We evaluate this field in the context of these criteria. First, however, we provide a summary of the MLP dataset and how it compares to standard GSE loan-level data available today. Standard GSE Data vs. Experian Mortgage Loan Performance (MLP) Data The MLP dataset contains thousands of fields describing mortgage performance from each borrower, loan, and property perspective, all refreshed monthly (including, amongst other things, new credit scores and refinance inquiry activity, loan performance, filed junior liens, and AVM values).  MLP differs from loan-level data provided by Freddie Mac, Fannie Mae, and Ginnie Mae, which the vast majority of market participants solely rely on, in a number of ways: Standard data provided by the GSEs and GNMA does not contain all the information necessary for accurate forecasting of mortgage prepayment and credit performance. Basic, critical fields like borrower’s current credit score and current junior liens on the property are missing. The new Mortgage Loan Performance (MLP) dataset from Experian contains borrower, loan, and property data fields covering the entire mortgage universe, including Agency, Non-Agency, and Esoteric mortgage products (CES, HELOC, Reverse), both securitized and non-securitized. MLP enables full three-dimensional (borrower + loan + property) tracking with persistent keys for borrower (before and after refinancing), loan (in securities/deals even after exit due to payoffs or buyouts, including before and after MSR sales), and property.  This enables end-to-end analysis of each borrower’s (and property’s) mortgage experience throughout their credit lifecycle. New, Material and Significant Field:  Student Loan Debt MLP contains a number of fields describing each mortgage borrower’s student debt load, including amounts in repayment, forbearance and collections; estimated interest rate, time remaining until forbearance expiration, and more. In the interest of simplicity, for this article we’ll focus on a single student loan-related field within MLP: Student Loans Balance, which is defined as the total balance on open non-deferred student trades reported in the last 3 months. Is Information Regarding Student Loans New to Markets? Standard loan-level data disclosed by the GSEs and GNMA contain no student-loan-specific fields. Theoretically, fields related to DTI at origination might capture some aspect of student loan debt. So, in the best-case scenario for an investor relying solely on standard disclosure, a DTI value as of origination is provided -- yet is never updated as the loan seasons and the borrower’s debt and income change (see more here).  But in the case of federal student loan debt attached to mortgages originated from early 2020 to late 2023, the level of detail provided by disclosure may be even more unknown due to COVID-era repayment and reporting moratoriums. The student loan repayment moratorium was a temporary federal policy that paused required payments, set interest rates to 0%, and suspended collections on most federally-held student loans. The moratorium began in March 2020, with payments resuming in October 2023, making it approximately 3.5 years in duration—the longest consumer credit payment pause in U.S. history. (Source: NCUA ) During the moratorium, student loan-related debt loads may have been understated as federal loans were in a temporary state of $0 repayment.  As an alternative to leaving student loan debt completely out of DTI calculations, an imputed payment equal to only 0.50% of the outstanding balance was often used as a placeholder for a borrower’s DTI calculation. As the Wall Street Journal recently reported, a new report from the Federal Reserve of New York shows a rise in student loan default rates by age group.  Student loan delinquencies have been broadly trending higher across all age groups.  Also, the average age of a borrower in default has risen to 40, and borrowers age 50 and older are now at a higher risk of default than younger groups. This 40 to 50-year-old age group represents prime home ownership years. Defaulted borrowers are also struggling to make other debt payments, too. The same report stated that almost 40% of past due student loan borrowers with auto loans are past due, 56% have at least one credit card past due, and 20% have a past due mortgage.  In addition to increased delinquency risk on their mortgage, borrowers with student loan debt also have fewer mortgage refinance options, as their elevated DTI may prevent qualification for a refinance, or increase the offered rate of a refinance and thereby reduce their incentive.  These dampening effects of student loan debt on mortgage CPR are clearly evident in the data, as described further below. Standard data only reports information related to the primary mortgage and does not include any details on the borrower’s other debts, with the exception of DTI at origination, which is never updated throughout the life of the loan. In contrast, MLP provides a comprehensive view of the borrower’s full credit profile, including other obligations such as credit cards, mortgages on other properties, student loan balances, and much more. Is Student Loan debt material to the residential mortgage market? Approximately $11 trillion of residential mortgage loans were originated during the student loan payment moratorium (Source: Experian MLP Dataset), a period marked by historically low mortgage rates during the COVID era.  As discussed above, DTI data contained in standard market disclosure may be particularly inaccurate for these loans.   As the Wall Street Journal recently reported, a new report from the Federal Reserve of New York shows a rise in student loan default rates by age group.  Student loan delinquencies have been broadly trending higher across all age groups.  Also, the average age of a borrower in default has risen to 40, and borrowers aged 50 and older are now at a higher risk of default than younger groups. This 40 to 50-year-old age group represents prime home ownership years.  Defaulted borrowers are also struggling to make other debt payments, too.   The same report stated that almost 40% of past due student loan borrowers with auto loans are past due, 56% have at least one credit card past due, and 20% have a past due mortgage.  In addition to increased delinquency risk on their mortgage, borrowers with student loan debt also have fewer mortgage refinance options, as their elevated DTI may prevent qualification for a refinance, or increase the offered rate of a refinance and thereby reduce their incentive.  These dampening effects of student loan debt on mortgage CPR are clearly evident in the data, as described further below.  Of today’s $13 trillion in outstanding mortgage debt, more than 10% of that debt ($1.5 trillion) is associated with borrowers who carry student loan debt.  For these borrowers, the average amount of student loan debt outstanding is approximately $50,000, versus a mortgage balance of approximately ~$289,000. In other words, the average student loan debt balance is almost 20% of the mortgage balance for the average borrower who carries both. For this set of borrowers, the average monthly payment is approximately $400 for student loan vs. approximately $2,200 for 1st lien mortgage—so that monthly student loan payments are a significant debt load, approximately 20% of the monthly mortgage payment.  (Source:  Experian MLP Dataset)  Is the effect of student loan debt a significant driver of performance? Figure 1 illustrates prepayments by student loan balance for a sample of loans drawn from MLP. The chart illustrates that borrowers with larger student loan balances prepay much more slowly, likely because some are effectively locked out of refinancing once student loan payments resume due to elevated DTI. The debt-to-income (DTI) ratio calculated using actual student loan payments may be significantly higher than the DTI calculated during the moratorium, in some cases exceeding GSE eligibility thresholds. As illustrated in Figure 1, for in-the-money (ITM) collateral, the differential between loans with material student loan balances (greater than $200,000) and loans with no student debt can reach up to 5 CPR. Notably, even for out-of-the-money (OTM) collateral, loans with student debt prepay 1 to 3 CPR slower, likely reflecting reduced mobility due to tighter financing constraints when purchasing a new home. Pools with otherwise similar prepayment characteristics may exhibit different prepayment behavior depending on the distribution of student loan exposure within their collateral. In addition, because loans with student debt tend to prepay more slowly, this effect increases over time due to burnout: loans without student debt prepay and exit the pools more quickly, leaving a higher concentration of slower-paying loans behind.  Given that 10% of the $13 trillion outstanding mortgage market is associated with borrowers who have student loans (Source:  Experian MLP dataset)—and that student loans have a meaningful impact on prepayments—many pools issued between March 2020 and October 2023 may be subject to this student loan debt CPR throttle, and therefore mispriced by investors relying exclusively on standard market data. Fig 1. Prepayment S-Curve: Student Loans Balance Source:  Experian MLP dataset hosted on IVolatility Data-Driven Platform _____________________________________________________ Michael Pyatski advises MBS traders, portfolio managers, quants, risk managers, loan originators, and technology professionals on making informed, data-driven business decisions that drive revenue growth, enhance risk management, and reduce trading costs. With more than 15 years of experience as an Agency RMBS trader—including serving as Head of the Proprietary Trading Desk at BNP Paribas—Michael developed and successfully implemented relative-value, data-driven profitable trading strategies to capture market opportunities embedded in data but not fully priced by the market. His trading experience, combined with a Ph.D. in econometrics, led him to found the Data-Driven Portal (https://datadrivenportal.com/), a platform that provides advanced technology for MBS trading and risk management. The platform’s No-Model Data-Driven technology leverages big data, econometric analysis, and AI to help traders identify relative-value opportunities in RMBS markets and generate above-market, risk-adjusted returns. _____________________________________________________

Published: June 17, 2026 by Perry DeFelice, Michael Pyatski