[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [cti-stix] Opinion Object Proposal
Terry:
This Opinion TLO proposal brings up a good point that I've been wanting to mention to the list. In traditional Intelligence there are well developed analytic techniques for discerning if a target or person of interest is using deceit and deception strategies as part of their mission. In cybersecurity intel we see this through the use of Avatars or Sock Puppets by Identities (Threat Actors), traversal through TOR and i2P, and through multi-layered fraud schemes like the use of money mules for money laundering operations. So in our field we could characterize the practice as denial, deceit, and obfuscation (DDO).
As an analyst trying to make sense of the expanded STIX data model as it is evolving I've been trying to imagine how I would characterize DDO tactics for a trust partner I'm sharing intel with.
Do I use the Identity (Threat Actor) TLO for an Avatar? How do I assert that I believe it is a persona? Do I use the CybOX Network Object for a TOR exit node characterzation? Do I need to assert any deductions I make about any connections to an Identity (Threat Actor) operating though TOR in this instance? And for the money mule example, I'm scratching my head on how to characterize that. Perhaps that is something we can drill down into when we get back to the TTP object post MVP release.
With an Opinion TLO I can make assertions about my interpretation of various Identities, TTPs and other features that are part of a natural process when I am hunting for clues and climbing up the pyramid of pain.
I strongly support the addition of an Opinion TLO for the Winter release, for all of the reasons noted by Allan and Bret.
Jane Ginn, MSIA, MRP
Cyber Threat Intelligence Network, Inc.
jg@ctin.us
Yeah this is a good point, I agree with waiting.
From:
<cti-stix@lists.oasis-open.org> on behalf of "Jordan, Bret" <bret.jordan@bluecoat.com>
Date: Friday, June 24, 2016 at 12:37 PM
To: Allan Thomson <athomson@lookingglasscyber.com>
Cc: Terry MacDonald <terry.macdonald@cosive.com>, "cti-stix@lists.oasis-open.org" <cti-stix@lists.oasis-open.org>
Subject: Re: [cti-stix] Opinion Object Proposal
Great catch Allan. Yes, we need to do originator confidence before we do third party confidence.
Bret
Sent from my Commodore 64
On Jun 24, 2016, at 7:32 AM, Allan Thomson <athomson@lookingglasscyber.com> wrote:
For the same reason that other confidence information has been targeted post-MVP, I would say this proposal should wait until beyond the MVP.
For example, there is *no* conveyance of confidence information currently in MVP by the original source of the intel but this proposal would be added so *other* people’s opinion’s would be conveyed?
That makes no sense.
allan
From: "cti-stix@lists.oasis-open.org" <cti-stix@lists.oasis-open.org> on behalf of "Jordan, Bret" <bret.jordan@bluecoat.com>
Date: Friday, June 24, 2016 at 5:41 AM
To: Terry MacDonald <terry.macdonald@cosive.com>
Cc: "cti-stix@lists.oasis-open.org" <cti-stix@lists.oasis-open.org>
Subject: Re: [cti-stix] Opinion Object Proposal
I love the idea, but I do not think we should do this for the Summer release. I see this being done in the Winter release.
The reason for that is this functionality is dependent on digitally signing content and that will NOT be ready for summer. Right it is being tracked as one of the primary things for the Winter release.
Thanks,
Bret
Bret Jordan CISSP
Director of Security Architecture and Standards | Office of the CTO
Blue Coat Systems
PGP Fingerprint: 63B4 FC53 680A 6B7D 1447 F2C0 74F8 ACAE 7415 0050
"Without cryptography vihv vivc ce xhrnrw, however, the only thing that can not be unscrambled is an egg."
On Jun 23, 2016, at 17:38, Terry MacDonald <terry.macdonald@cosive.com> wrote:
Hi All,
Can I take it that the lack of responses means that you all think this is a great idea? If so that's excellent, as it means it can drop straight into the MVP build as it doesn't require any modification :).
Though seriously, if everyone is OK with the idea of this object which I've been banging on about for about a year then please speak up so we can get it added and allow people to have opinions about other's assertions. This object opens up the ability for people to effectively 'upvote' or 'downvote' a piece of threat intelligence. This will allow consumers to crowd-source how much they should trust the assertions made in that threat intelligence - which is a key enabler for consumers to effectively use the threat intelligence they receive.
I passionately believe we need this object in MVP.
Use Case (bad intel):
- Threat Intel Vendor A provides some high confidence threat intel saying that 8.8.8.8 (Google DNS) is a malicious asset.
- 30 other vendors, producers and generate Opinion objects that all strongly disagree with the intel that Vendor A released.
- A consumer can now see that Vendor A's intel shouldn't be trusted to have a high confidence, and therefore shouldn't probably be used in production.
OUTCOME: Confidence in the value of the threat intel is decreased
Use Case (good intel):
- Threat Intel Vendor B provides some low confidence threat intel saying that they think that www.compromisedsite.com has been compromised by Angler.
- Threat Intel Vendor C sends an Opinion Object strongly agreeing with Threat Intel Vendor C as they believe they are correct
- A consumer can now see that Vendor B's intel is pretty good, and they can potentially increase their confidence in that intel, and maybe use it in production.
OUTCOME: Confidence in the value of the threat intel is increased
What say you STIX community?
Cheers
Terry MacDonald | Chief Product Officer
<cosive_mail_signature.png>
On Thu, Jun 16, 2016 at 11:00 PM, Terry MacDonald <terry.macdonald@cosive.com> wrote:
Hi All,
As I've mentioned many times over the last year I firmly believe we need a way for third parties to agree or disagree with the threat intelligence they have received. If Org A has released a high confidence relationship between ActorX and Campaign G, and Org B knows that the relationship is wrong, then they need a way of signalling that to the community, so that community members don't blindly accept what Org A has released.
Since late last year I've been suggesting we need an Opinion object. And today I took the step of writing up what that would look like.
I would like to propose that we add this to the draft as proposal, and that we include it in the MVP release.
1.2.Opinion
Type Name: opinion
Status: Proposal
MVP: Undecided
The Opinion object is used to convey the Object creator's opinion about another object produced by a third-party. It will allow each organization to agree or disagree with another organization's assertions, and ultimately will enable consumers to collect and understand the collective opinions of the community about the quality of the threat intelligence they have received.
This is the first step towards consumers being able to crowdsource the opinion of the community, which will help newcomers to the threat intelligence sharing groups better understand which threats have a high degree of third party agreement and which are contentious.
1.2.1. Properties
STIX TLO Common Properties
type, id, created_by_ref, revision, created_time, modified_time, revoked, revision_comment, object_markings_refs, granular_markings
Property Name
Type
Description
type (required)
string
The value of this field MUST be opinion
description (optional)
string
A description that provides the recipient with reasoning to back up the opinion identified in this Opinion object.
object_ref(required)
identifier
The id of the object that the Opinion refers to. This id can be any other STIX TLO except another Opinion object.
opinion(required)
list of type controlled-vocab
The opinion that the producer has about the object listed in the object_ref field. This is one of the following options:
? "strongly-agree"
? "agree"
? "neutral"
? "disagree"
? "strongly-disagree"
? "no-opinion"
1.2.2. Source Relationships
These are the relationships defined between the Opinion Object and other objects.
STIX TLO Common Relationships
duplicate-of, related-to
1.2.3. Destination Relationships
These are the relationships defined between other objects and the Opinion Object.
Kind of Relationship
Source Type
Description
evidence-of
observation
Relates the Observation to an Opinion providing the evidence that the opinion was based on. This observation is evidence of why the organization formed the opinion it did about the threat intelligence contained within the object_ref field.
Cheers
Terry MacDonald | Chief Product Officer
<cosive_mail_signature.png>
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]