DSS

DSS 2026 / Day 1 / Pearl Hall

One Year Later: Surviving the SOC 2 Type 2 Observation Window in DeFi

Roxanne Bartnos

17:26 · Recording

Watch on YouTube ↗Open recording viewerOriginal scheduled programme

Roxanne Bartnos presented this recording in place of the originally scheduled speaker. The linked programme preserves the original schedule.

Download plain textDownload timestamped JSON

Source: YouTube automatic caption track · Exported 2026-11-16. Timestamps link to the source video. Cue end times were not supplied.

Transcript

0:01Hi everybody. I'm Roxanne Bodnos from

0:04Zero Shadow. I head up risk and security

0:07management

0:09focusing more on the proactive security

0:12side at Zero Shadow.

0:15So

0:16I was supposed to be speaking on with my

0:19client

0:20and Pierre Yovan from 1 inch.

0:22Unfortunately, he wasn't able to make

0:24it, so I will be taking over.

0:28So we will be focusing on

0:31assurance, SOC 2, ISO 27001

0:36and and how it works in terms of the

0:39framework that Zero Shadow has

0:42developed, security risk management.

0:46So when a business achieves ISO 27001 or

0:50completes its SOC 2 audit,

0:53it's an important milestone.

0:57But what happens next?

1:00What happens the next day, the next

1:01week, the next month?

1:03The risks are still there.

1:05The business is still changing and

1:07people still have to operate those

1:09controls.

1:10That is what I want to talk about today,

1:13security risk management or SRM, a

1:16security risk management framework that

1:18Zero Shadow has developed.

1:21And I want to talk about how our

1:22partnership with 1 inch developed from

1:25understanding their exposure into

1:27embedding security in the way they work.

1:40Right. So there's been a lot of trending

1:42posts on LinkedIn regarding Binance

1:45um and I thought that Binance the

1:48Binance incident brings this

1:49conversation into focus.

1:52Um here is an organization that has

1:54publicly announced

1:56SOC 2 Type 2 and ISO credentials and has

2:00reported a serious security incident.

2:05That does not mean the standards are

2:08worthless or that the incident proves

2:10Binance treated compliance as a

2:11checkbox.

2:13What it does remind us is that insurance

2:16or assurance cannot guarantee that an

2:19organization will never get breached. It

2:22provides confidence about specified

2:25aspects of a program. It cannot remove

2:28every exposure or anticipate every

2:31change.

2:33The problem

2:34for any organization

2:39is achieving

2:41or the problem for any organization is

2:43when achieving the report becomes the

2:45objective. So they work towards SOC 2

2:48attestation or their ISO 27001

2:52certificate and they think, hey, the

2:54job's done.

2:56But this is where the work behind then

2:58loses attention afterwards.

3:00A control still needs to operate when

3:03someone joins, when application changes,

3:06or when an unusual transaction appears.

3:09That is the connection between SRM.

3:13We keep asking what could hurt the

3:16business, whether our response is

3:18working, and whether our people are

3:21prepared.

3:22So let me bring that back to how the

3:25journey with 1 inch started.

3:32The first step that we took with 1 inch

3:35with the partnership was developing a

3:37shared understanding of 1 inches risk

3:40exposure.

3:42Before we could build a useful strategy,

3:45we needed to understand where the

3:48attention was required.

3:50Now that meant looking beyond a list of

3:53technical controls

3:55because technology is just part of it.

3:57The systems, access, and infrastructure

4:00that supports the business, that's just

4:02a a single part.

4:04But also what we need to understand is

4:06who operates those systems, how

4:09ownership works, and what happens when

4:11an issue needs a decision.

4:14Then there is the operational picture.

4:18What visibility does the organization

4:20have?

4:21How do you or how do concerns reach the

4:24right people?

4:25What would happen if a routine issue

4:29became an incident?

4:31The assessment that we did with 1 inch

4:33gave us a starting point for those

4:35conversations.

4:37It helped us move from a long list of

4:41possible improvements to a shared view

4:43of what matters most.

4:47That was the foundation of our journey.

4:49Understanding the exposure, recognizing

4:52the gaps, and agree where to act. The

4:54security strategy could then develop

4:58around the realities of the business

5:00with a clear reason behind the work we

5:03were asking people to do.

5:11Sorry, I think I need to go one back.

5:18All right, so

5:20once we understood the exposure, we then

5:23started prioritizing.

5:25We knew we could not treat every gap as

5:28equally urgent, and completing the

5:30easiest tasks would not necessarily

5:33address the most important risks.

5:36We focused attention on the areas that

5:39had the greatest potential to impact the

5:41business. Then worked through how those

5:44risks could be mitigated. That meant

5:47agreeing what needed to change, who

5:50could make it happen, and how would we

5:53know the action has addressed the risks

5:55associated with that control.

5:58Now as the work progressed, the strategy

6:01became more informed. We were able to

6:04connect

6:05immediate improvements to the longer

6:06term direction, adding value right from

6:09the start,

6:10rather than dealing with each

6:12requirement as isolated tasks. So we

6:15removed the siloed approach and we took

6:17a holistic approach.

6:19In 1 inch's own account, we saw the

6:21operational areas that needed attention.

6:24Um software as a service, access

6:26removal, cloud resilience and

6:28configuration,

6:30um endpoint enforcement and clearer

6:32control on ownership across the

6:35business. Now these are examples from

6:37their journey.

6:39This is SRM in practice. This is what

6:42security risk management does. The

6:44exposure gives us the reason to act.

6:47Prioritization gives us the order.

6:51Ownership gives us a way to get the work

6:53done. But even a well-designed technical

6:57team or technical control depends on the

7:00people expected to run it.

7:08Right. So

7:11this is the strongest example in 1

7:14inch's journey.

7:16Policy acknowledgement moved from 47% to

7:19100% of employees in scope.

7:22The interesting part is what changed

7:24underneath that number.

7:26When the process relied on Slack

7:28reminders and goodwill, people still had

7:31competing priorities.

7:34Once it became systematic, the

7:36organization had a clearer way of making

7:39the task happen.

7:41There was one source of truth, monthly

7:43reconciliations between HR and

7:46compliance records helped keep the

7:47population accurate. But the 1 inch team

7:51built automated nudges. They named

7:53owners and leadership escalation meant

7:56outstanding actions had somewhere to go.

8:02That is how 1 inch started changing

8:03habits. They started changing culture.

8:06They made their expectation clear, made

8:09the action part of the routine and gave

8:11people the support and accountability to

8:14follow through.

8:16Acknowledging a policy does not on its

8:18own prove that someone understands it or

8:22behaves securely, but the example shown

8:25shows how an incomplete process became a

8:28repeatable process.

8:30The same thinking applies to the people

8:32running access reviews, approving

8:34changes, or reporting incidents. The

8:36first step is intent, understanding the

8:38intent of the organization.

8:41Yvonne and the broader One H team

8:43deserve the credit for embedding that

8:45change. Zero Shadow guided and

8:48challenged the program, but their

8:51leadership and teams inside the business

8:53had to make it work.

9:01So

9:03we know that preparedness starts with

9:05policy acknowledgments, understanding

9:07the intent of the organization. So how

9:10do we take this further?

9:13Preparedness is where we look at whether

9:16people can act when the situation stops

9:19being routine.

9:20Having a document available is only part

9:22of that.

9:24We need to ask

9:25can people recognize the the signals

9:28that matter?

9:30Do they understand what needs to be

9:31reported and where to raise it?

9:34Once something is reported, who can make

9:37the decisions? The team needs to

9:39understand its authority, escalation

9:42paths, and the actions it can take to

9:45contain the problem.

9:47Then we need to practice, and exercise

9:50helps reveal whether the people

9:52understand the responsibilities and

9:54whether the response works

9:58with the systems and dependencies

10:00dependencies they actually use.

10:03Finally, we need to learn findings from

10:06exercises, audits, and real incidents

10:09should influence priorities and

10:11controls.

10:12Otherwise, we keep producing reports

10:14while repeating the same weaknesses.

10:17Now, those are the SRM operating

10:19principles. The point is that

10:22preparedness has to connect the plan to

10:25the people expected to act.

10:27Your people are your biggest assets, but

10:29also can be your greatest weakness.

10:32Their understanding, judgment, and

10:34ability to work together are part of the

10:36organization's

10:38resilience.

10:42So, let's talk about internal audits

10:46and how expectation met operating

10:50reality.

10:52Zero Shadow conducted the first internal

10:54audit for

10:561 inch, bringing into focus 1 inch's

10:58expectation versus the operating

11:00reality.

11:02The observation window that we had

11:05brought those lessons into focus.

11:07These are the themes that came through

11:09in 1 inch's own account.

11:12One's expectation

11:14One expectation was that you can fix the

11:17gaps before the audit. The reality was

11:19that the controls still need to operate

11:22throughout the period

11:24cuz when a failure appears, someone

11:26needs to see it, understand it, and

11:29understand its significance in order to

11:31act.

11:33Another expectation that the team had is

11:36that security will drive compliance,

11:39but security cannot perform every

11:41control across HR,

11:43engineering, and the rest of the

11:45business. Each team needs to understand

11:48their own part

11:50in the ownership and what they

11:52contribute to security um in the

11:55organization.

11:56Then there's the assumption that strong

11:59technical controls will be enough,

12:02but those controls still depend on

12:04people making the right decisions and

12:07performing the right actions

12:09consistently.

12:11This is what embedding

12:13security into your organization means.

12:16The control becomes part of how the

12:18organization works and the evidence

12:20reflects that it works.

12:22That was an important shift in the

12:24journey.

12:25Less dependence on someone chasing an

12:27audit task and more ownership of an

12:30ongoing responsibility.

12:35So the assurance milestones do matter in

12:38every organization. One Inch achieved

12:40ISO 27001 certification in 2025,

12:44maintained it through its 2026

12:46surveillance audit with zero

12:47non-conformities and reached its SOC 2

12:50milestone in 2026.

12:53Zero Shadow supported that work through

12:56activities including the internal audit,

12:58head of surveillance, and participation

13:00in auditor sessions, but the outcome

13:03depended on the program the One Inch

13:05team was operating.

13:07It is also important to be precise on

13:10what is SOC 2 and what is ISO 27001

13:13about.

13:14SOC 2 Type 2 examines control design and

13:17operating effectiveness over a defined

13:20scope and period. It produces an

13:22attestation report, whilst ISO 27001

13:26addresses a management system that

13:28already includes risk management and

13:30continual improvement.

13:32So security risk management belongs

13:34within that picture. It keeps the

13:37underlying disciplines active as the

13:40business changes. The issue with the

13:42tick box approach is allowing the report

13:44to become the finish line.

13:46For us, the more useful question is, can

13:49the organization sustain the work that

13:51supports the assurance and can its

13:54people respond when something happens

13:57despite the controls?

14:01So where is One Inch today in their

14:02journey? Their journey currently

14:04reflects an ongoing

14:07proactive approach to security. They

14:09have reached important assurance

14:10milestones and made parts of their

14:12operating environment more systematic.

14:15The work continues as the organization

14:17changes.

14:18Our partnership extends through

14:20assurance support and operational

14:21security work.

14:23The broader SRM responsibility remains

14:26active, continuously reviewing risk

14:29exposure, assessing whether the

14:32treatments we have in place are working

14:33and respond when risks emerge. That

14:37requires ownership inside One Inch

14:39alongside advice and challenge from Zero

14:42Shadow.

14:43Now this also supports the business

14:45conversation. A TradFi partner wants to

14:48understand who is accountable, how risk

14:51is managed, and what evidence supports

14:53the answer.

14:54A functioning program gives One Inch a

14:57stronger basis for the discussion. It

14:59also creates a foundation for assessing

15:02the additional requirements of new and

15:04emerging markets. That does not

15:06guarantee market access or mean a breach

15:08cannot happen.

15:10It means the organization is better

15:12placed to explain and manage its

15:14exposure.

15:15You can see the value of that ongoing

15:18work in how One Inch describes the

15:20payoff.

15:23So this is what they look at it from

15:25this perspective. Their due diligence

15:27conversation became easier to support

15:30with an established certificate,

15:32attestation report, and policy pack.

15:35Structured vendor management helped with

15:37assess suppliers and respond when others

15:40assessed them. They also described

15:43greater confidence from investors and

15:45leadership and an ability to strengthen

15:47their posture while retaining their DeFi

15:50operating culture.

15:51That is the commercial connection to

15:53SRM. A functioning program gives the

15:56business a more credible way to explain

15:58its risk management while helping the

16:01team operate more consistently. So let

16:03me close with the four lessons from

16:05their journey.

16:12First, SOC 2 Type 2 adds examination of

16:14operating effectiveness over the period.

16:17The important idea for our discussion is

16:19that the controls need to work as part

16:21of normal operations. The broad

16:24statement about 2026 losses is one

16:26inches framing rather than a claim that

16:29an audit predicts or prevents every

16:31exploit.

16:32Second, compliance can help expose

16:34weaknesses and establish useful habits

16:37when the work stays connected to the

16:39underlying risk.

16:40Third, culture needs visible ownership.

16:43The policy acknowledgement example shows

16:46the difference a systematic approach,

16:49automation, and leadership escalation

16:51can make.

16:52Finally, a DeFi business can retain its

16:55operating culture while developing the

16:58assurance its institutional partners

17:00need.

17:01Bring that back to our starting point.

17:03The journey began with understanding

17:04exposure and prioritizing the gaps. It

17:08developed through strategy, controls,

17:10and people's habits with assurance

17:12marking progress along the way. I will

17:15end with saying assurance is a

17:17milestone. Security risk management is

17:20what the organization does every day.

17:23Thank you.

Source and accuracy details

Source export SHA256: 6ac21583a96e8236ee084e28826f5f1d1c4e61b5427953fcff12067492b24b6a