Situational Awareness Terminal
◈ Source Credibility Index
1. BLUF (Bottom Line Up Front)
An unidentified actor injected malicious JavaScript into scripts served by Optimizely Web on 16 November 2026, and visitors to 178 organizations, including CNN, the BBC, HSBC, Equifax, Zoom, Salesforce and IKEA, received it. Available reporting best supports a financially motivated, multi-purpose campaign (address swapping, card skimming, cookie and screenshot theft) with publicly documented iPhone exploit kits attached. It is likely (about 80% confidence) that the direct financial loss was small, roughly $7,000 on-chain, and that Optimizely has cleaned the scripts. Not reported: how many cards, cookies or iPhones were compromised, and who the attacker is.
2. Key Judgments — Optimizely Supply-Chain Injection Campaign
- The compromise is established. Report URI's Scott Helme found 188 infected scripts at 178 organizations among the 1,611 Optimizely scripts it checked. Optimizely confirmed the incident, according to IBTimes and Coinspeaker. Four outlets report it, but most trace back to Helme's write-up, so the facts are strong while independent verification is thinner than the outlet count implies.
- The on-chain theft is small. IJR reports about $7,040. Coinspeaker cites Report URI at about $7,150 across Bitcoin, Litecoin, TRON, Ethereum and BNB Chain, and its own check of the Bitcoin wallet shows about $5,500 in five payments, one of them 89% of the total. Card data, cookies and screenshots have no reported loss figure, and Salesforce Ben says the full extent of damage is unknown.
- The iOS component is a capability and not a confirmed outcome. The code loaded Coruna (iOS 13–17) or DarkSword (iOS 18) through a hidden iframe, and iOS 26 devices were only fingerprinted. IBTimes says it was not clear whether any device was compromised, and Reuters could not confirm successful use.
- The intrusion mechanism is only partly explained. Optimizely, per BleepingComputer via IBTimes, cited a flaw in its account signup process. Salesforce Ben, relying on Tokenpost, says it was unclear how attackers gained write access to customer accounts. Both agree the code went into a project JavaScript field and was served from Optimizely's own delivery network.
- The actor is unattributed. Helme told the Daily Caller News Foundation (via IJR) that the same attacker has been active since at least March, which points to an earlier period of activity that has not been described.
3. Analysis of Competing Hypotheses (ACH)
| Hypothesis | Supporting Evidence | Contradicting Evidence | Evidence Gaps | Probability |
|---|---|---|---|---|
| H-A: Financially motivated criminal actor running an opportunistic, multi-tool campaign, with off-the-shelf iOS kits added to widen yield. | Wallet swapping across seven chains, card skimming, and cookie and screenshot theft all point to monetization. Wallets were listed and the attacker emptied the Bitcoin wallet at 05:39 UTC on 17 November. Coruna and DarkSword were documented earlier in the year by Google, iVerify and Lookout (IBTimes), so no bespoke exploit work is reported. | Yield of about $7,000 is low for a campaign reaching 178 large sites. Hosted payment fields such as Stripe were out of reach (Coinspeaker). The same actor reportedly operated since March. | Total card records captured, cookie reuse, funds flow after the wallet exit, and the March-onward activity. | 55% |
| H-B: The main objective was device access through the iOS exploit chains, with crypto and card theft secondary or cost-offsetting. | Version-specific exploit delivery via boot.js, and an iOS 26 fingerprinting stage that may be a probe for a later chain (Salesforce Ben). Screenshot and cookie collection also serve access goals. | No report of any compromised device. Coruna is described as ineffective against current Apple software. The iOS 26 stage only gathered device information. | Payload behavior after exploitation, victim telemetry from Apple or security vendors, and the kits' command infrastructure. | 27% |
| H-C: Access came through customer-account or credential abuse more than a platform-level flaw (an intrusion-mechanism hypothesis that can coexist with H-A or H-B). | Code was written to a customer-project field. Salesforce Ben says the access path was unclear. | Optimizely, per BleepingComputer via IBTimes, attributed the intrusion to a flaw in its account signup process and said it fixed it. | Optimizely's technical post-mortem, the number of accounts abused, and how accounts were created or taken over. | 12% |
| H-D (Maskirovka / Strategic Deception): The apparent signal is a deliberate disinformation, fabrication, or denial-and-deception operation designed to shape perception or mask a different course of action. | Small theft against large reach could suggest a decoy. The $7,000 figure "could not be independently verified" (IBTimes). Some victims have not commented. | Report URI, Optimizely, Salesforce and Equifax each acknowledge the event. Hudson Rock reported evidence independently. A timeline with UTC timestamps and public wallet addresses is consistent with real activity. | Whether the exposed wallets are the only ones, and whether the campaign masks a quieter data-collection aim. | 2% |
ACH Assessment: H-A is best supported because the observed behavior (swapping, skimming, public wallets, a quick exit) matches revenue generation. H-B cannot be excluded, because attackers would not need a version-aware iOS chain for the stated take. Available reporting shows no formal contradictions. The differences are partial: the iOS vulnerability threshold is given as 18.6 or earlier (IJR) versus patched at 18.7.3 and 26.3 (Coinspeaker, Salesforce Ben), and the entry vector is described differently. These reflect incomplete technical disclosure and do not materially weaken the core finding.
4. Key Assumption Check (KAC)
- Critical Assumptions: (1) Report URI's 188-of-1,611 sample captured the full set of affected scripts. If not, the victim count exceeds 178. (2) Cleanup was complete by 23:11 UTC on 16 November, as reported. Residual infected scripts would extend exposure. (3) The wallets listed by Report URI are all the attacker's. Additional wallets would raise the loss figure. (4) The iOS kits were mostly ineffective against patched devices. Evidence of successful compromise would raise severity. (5) Optimizely's signup-flaw explanation is accurate. If it is not, the stated fix may not close the access path.
- Information Gaps: The number of card records and cookies taken (needs victim and processor reporting). Whether any iPhone was compromised (needs Apple or vendor telemetry). The access method and the number of accounts abused (needs an Optimizely post-mortem). The attacker's identity and the nature of earlier activity since March. Whether Zoom, HSBC, CNN and Apple, who did not respond to Reuters, have findings.
- Bias & Deception Risks: Single-source echo is the main risk: outlets repeat Helme's analysis and the same dollar figure. Report URI is a security vendor with a commercial interest in publicity, though its data is specific and checkable. Coinspeaker is crypto-focused and carries promotional wallet and token content, and it frames the event partly through market impact. The Coinspeaker piece embeds a social-media post from the account "The Hormuz Letter" naming Fox News and Air France-KLM, which is unverified from that source alone. IBTimes attributes some points to Reuters, MacRumors and BleepingComputer, which this analysis did not view directly. Victim statements (Salesforce, Equifax) are self-reported. No adversary deception indicator is present.
5. Implications and Strategic Risks — Optimizely and Third-Party Script Users
Most of the remaining risk is delayed: stolen card numbers, cookies and screenshots can be used after the scripts were cleaned. The attack worked because trusted vendor-served code runs inside first-party pages, a pattern that also applies to analytics, advertising and testing tools. Follow-on disclosures from the 178 organizations and from Optimizely's technical review will determine whether the campaign is closed or still developing.
Cyber / Information Space — Optimizely Web and customer sites
The injected code was served from Optimizely's own servers, so standard allow-lists would not have flagged it. Salesforce disabled Optimizely scripts across preview and production, and Equifax stopped using the service, which shows how quickly customers can drop a vendor. Optimizely said it added protections against unauthorized code delivery but has not published detail.
Economic / Social — Banking, finance and media victims
Helme's report, via IJR, lists 32 banking and finance organizations (including Paychex, Equifax and SmartAsset) and 10 news and media organizations (CNN, Fox News, BBC). Customers of these sites who entered card details on 16 November face fraud exposure. Coinspeaker reports that Bitcoin moved 0.35% during the attack window, so no market effect is reported. Coinspeaker also cites DefiLlama data placing frontend and infrastructure attacks at 74% of 2026 hack losses, but most of that figure comes from a single Binance incident of $6.36 billion, so it describes the category more than this event.
Security — Apple iPhone and iPad users
Devices below iOS 18.7.3 or 26.3 were exposed to exploit delivery by merely loading a page, according to Report URI as cited by Coinspeaker and Salesforce Ben. The exploit kits remain publicly documented, so other actors can reuse them in separate injections.
6. Recommendations and Outlook
- Immediate Actions (0–30 days): Track Optimizely's post-mortem for the access method and the number of accounts affected. Monitor victim notices on card and cookie exposure. Watch the attacker's published wallets for further movement. Check whether other organizations that use Optimizely report additional infected scripts. Monitor for reports of confirmed iOS compromise tied to Coruna or DarkSword.
- Medium-Term Posture (1–12 months): Track whether affected firms adopt script-integrity controls (content security policy, subresource integrity where applicable) and whether vendors change account-creation and write-access controls. Watch for similar injections through other testing or personalization services, and for attribution links to the earlier activity Helme dates to March.
- Scenario Outlook: Most likely: Optimizely publishes a limited explanation, losses stay near $7,000 on-chain, and card-fraud reports are sporadic. Best: Optimizely's review shows the access path closed and no iOS compromise is confirmed. Worst: telemetry shows successful iPhone compromise or large-scale card and cookie reuse, or the actor re-enters through other accounts. Triggers are new victim disclosures, vendor telemetry, and fresh infected-script findings.
7. Key Individuals and Entities
| Name | Role / Affiliation | Relevance to Assessment |
|---|---|---|
| Unidentified attacker | Actor behind the injection; reportedly active since at least March | Attribution and objective unknown; wallets were publicly listed |
| Optimizely | Provider of the Optimizely Web experimentation platform | Confirmed incident; cites an account-signup flaw; said it fixed it, removed the code and notified customers |
| Scott Helme / Report URI | Researcher and founder of the security firm that found and published the campaign | Primary source for the 178-organization count, timeline and loss figures |
| Hudson Rock | Cybersecurity firm | Reported evidence that sites served malicious code (IBTimes) |
| Salesforce | Victim; own systems reportedly not compromised | Investigating; disabled Optimizely scripts; domains named include evergage.com, exacttarget.com, pardot.com and salesforce-scrt.com |
| Equifax | Victim | Stopped using Optimizely; found no evidence its systems were compromised (IBTimes) |
| Ledger | Hardware wallet maker; victim website | Script cleaned at 20:41 UTC; no sign of hardware wallet breach (Coinspeaker) |
| IKEA, Zoom, HSBC, CNN, BBC, GitLab, Paychex | Affected organizations | IKEA assessing impact; Zoom, HSBC and CNN had not responded to Reuters; BBC hosted the first infected script at 04:21 UTC |
| Apple | Device and OS vendor | Did not immediately respond to Reuters; patched iOS versions are the stated mitigation |
8. Thematic Tags
Cybersecurity, supply-chain-attack, Optimizely, web-skimming, crypto-clipper, iOS-exploit-kits, third-party-scripts, Report URI
Explore more: Cybersecurity Briefs · Daily Summary · Support us
✓ YES Dissemination
✓ Cleared Analyst review
| Source | SCI | Role |
|---|---|---|
| coinspeaker | 3 | SOURCE_DOCUMENT |
| salesforceben | 3 | SOURCE_DOCUMENT |
| ijr | 3 | SOURCE_DOCUMENT |
| ibtimes | 2 | SOURCE_DOCUMENT |