Washington built cyber sharing hubs in February 2015 and left the reading to understaffed recipients
In one week the White House announced a cyber intelligence center for policymakers and an order to multiply private sharing groups. Both moved information. Neither paid for the analysts at counties, small agencies and hospitals who had to act on it.
Looking Back revisits past weeks one at a time, up to the most recent, and asks what it meant for the banks, health systems and public agencies we serve. The series was written in 2026, with the benefit of hindsight, and each article says which week it looks back on.
Two-minute summaryThe dek and the first sentence of each section, quoted from the article
In one week the White House announced a cyber intelligence center for policymakers and an order to multiply private sharing groups. Both moved information. Neither paid for the analysts at counties, small agencies and hospitals who had to act on it.
- The week Two announcements in the week of February 9, 2015 changed the structure of federal cyber threat sharing.
- The producers: a small center for policymakers CTIIC was designed to fill a gap at the top of the chain.
- The brokers: the NCCIC and a new class of sharing groups The executive order addressed the middle of the chain: the organizations that pass information between government and everyone else.
- The consumers: where indicators become protection The end of the chain is the organization that receives a warning and has to act.
- The gatekeepers: privacy, legal and classification review Information sharing is constrained by rules about what may be shared.
- How the gap differed by type of organization The same chain looked very different depending on who sat at the consuming end.
- The objection: sharing structures had to come first The strongest objection is that the order was the right first step.
- With hindsight The structures announced in February 2015 mostly came into being.
The week
Two announcements in the week of February 9, 2015 changed the structure of federal cyber threat sharing. Both were made at public events, and both created new places for information to flow through.
On Tuesday, February 10, Lisa Monaco, the President’s Assistant for Homeland Security and Counterterrorism, spoke at the Wilson Center in Washington. She announced that the Director of National Intelligence would establish a Cyber Threat Intelligence Integration Center, modeled on the National Counterterrorism Center. She described a gap: no single government entity was responsible for producing coordinated cyber threat assessments, sharing information rapidly among the existing cyber centers, or supporting operators and policymakers with timely intelligence. The new center would integrate and analyze intelligence, but it would not collect intelligence itself or duplicate the work of other centers (White House, February 10, 2015).
The speech placed the center inside a broader strategy. Monaco cited recent attacks on retailers, a bank, Sony Pictures and the health insurer Anthem, whose breach had been disclosed the week before. She described a Cyber Response Group created the previous summer to coordinate interagency responses. She said the government had distributed information and malware signatures to the private sector within a day of the Sony attack. She also noted that the President’s new budget proposal included about 14 billion dollars for cybersecurity (White House).
On Friday, February 13, the White House held a Summit on Cybersecurity and Consumer Protection at Stanford University. There the President signed Executive Order 13691, Promoting Private Sector Cybersecurity Information Sharing (Executive Order 13691). The order had four main parts.
First, it directed the Secretary of Homeland Security to strongly encourage the formation of Information Sharing and Analysis Organizations, or ISAOs. They could be organized by sector, sub-sector, region or any other affinity, could be public, private or mixed, and could be for-profit or nonprofit. The existing Information Sharing and Analysis Centers, built around critical infrastructure sectors since the late 1990s, were one model. The order wanted many more, for communities the sector model did not cover.
Second, it told DHS to select, through an open competition, a nongovernmental ISAO Standards Organization. That body would identify voluntary standards and guidelines for creating and operating ISAOs, covering contracts, business processes, operating procedures, technical means and privacy protections such as data minimization.
Third, it made DHS’s National Cybersecurity and Communications Integration Center, the NCCIC, a critical infrastructure protection program under the Critical Infrastructure Information Act of 2002. The NCCIC could enter voluntary agreements with ISAOs, and DHS could determine whether ISAOs and their members were eligible for security clearances. Amendments to the National Industrial Security Program order let DHS make those determinations for contractors and grantees in such a program.
Fourth, it required agencies to build privacy and civil liberties protections into their sharing activities, based on the Fair Information Practice Principles. Agencies were to report assessments to DHS’s Chief Privacy Officer and its Office for Civil Rights and Civil Liberties. The order stated that it gave no agency new authority to regulate critical infrastructure security (Executive Order 13691).
Two weeks later, on February 25, the White House published a fact sheet and presidential memorandum formally establishing the new center, known as CTIIC. The fact sheet said it would be small, with about 50 government personnel drawn from relevant departments and agencies. It would support the NCCIC, the FBI-led National Cyber Investigative Joint Task Force and U.S. Cyber Command. It would not be an operational center, would not manage incident response and would not engage private-sector entities directly (White House fact sheet, February 25, 2015).
The week built on January’s legislative proposals, which had asked Congress for liability protections for companies that shared threat indicators. The executive order did what could be done without legislation. It shaped the organizations that would carry information between government and everyone else.
February 2015 is remembered for new structures: a federal intelligence center and an order to grow private sharing groups. Both were designed to move information. Both left untouched the step where information becomes protection, which is the work of analysts and engineers inside the receiving organizations. For federal agencies with security operations centers, that was manageable. For counties, school districts, small agencies and community hospitals, it was the binding constraint, and the new structures did not address it.
Threat sharing is a chain of work, and each link needs people. We take the links in order, one section each. The first is the producers, who turn raw intelligence into assessments. The second is the brokers, who move information between government and other organizations. The third is the consumers, who act on what arrives. The fourth is the gatekeepers, who decide what may be shared at all. Those four roles do not overlap and together cover the chain. Two shorter sections follow: one on how the staffing gap differed by type of organization, and one testing the argument against its strongest objection.
The producers: a small center for policymakers
CTIIC was designed to fill a gap at the top of the chain. Before it, the FBI, the National Security Agency, DHS, Cyber Command and other agencies each produced their own views of cyber threats. No one was charged with combining them into a single assessment for the President and senior officials (White House).
What the center was built to do
The fact sheet described an all-source analysis center. It would integrate intelligence on foreign cyber threats, support the existing cyber centers and help coordinate government efforts to counter threats (White House fact sheet). Its model, the National Counterterrorism Center, had been created after the September 11 attacks to fuse terrorism intelligence across agencies. The cyber version was much smaller and had no collection authority of its own.
Its customers were inside government: the National Security Council, the Cyber Response Group and the operational centers. When a major incident occurred, CTIIC’s job was to give policymakers a coordinated account of who was responsible and what they were likely to do next. The Sony attack, attributed to North Korea within weeks, was the recent example of a case where officials had to assemble that picture quickly.
What it was not built to do
The fact sheet was explicit about the limits. CTIIC would not engage the private sector directly, would not respond to incidents and would not direct investigations (White House fact sheet). Whatever it produced would reach companies and state and local agencies only indirectly, through the NCCIC, the FBI and other channels.
Those limits were sensible. A new intelligence center talking directly to companies would have duplicated DHS and the FBI and raised civil liberties concerns. They also meant that CTIIC added nothing to the capacity of the organizations that would in the end have to defend themselves. A better-informed National Security Council could make better decisions about sanctions, indictments and diplomatic responses. It could not patch a county’s servers.
Fifty people and the analyst market
About 50 staff, drawn from existing agencies, was a modest commitment (White House fact sheet). Those people came from the same small pool of experienced cyber intelligence analysts that the operational centers, the military and the private sector were all trying to hire. Detailees to a new center left gaps in their home agencies. The integration center reorganized existing expertise more than it added to it.
This mattered for the rest of the chain. In 2015, experienced threat intelligence analysts were scarce across the economy. Banks, defense contractors and technology firms were building their own intelligence teams and paying well for them. Federal pay scales and the time required for security clearances made government recruiting harder. Every structure announced that week assumed analysts who would have to come from somewhere.
Attribution and what it is for
Much of CTIIC’s intended value lay in attribution: establishing who was behind a major intrusion with enough confidence to support a response. Attribution informs sanctions, indictments and diplomacy. For a defender, it helps mainly by indicating what an adversary is likely to want and how it usually operates, which can guide where to look. That benefit, too, reaches defenders only when someone translates it into specific behaviors to watch for. A finding that a foreign service was behind a breach is useful to a county only if it comes with guidance the county’s staff can apply.
What a policy center can and cannot change
An integration center improves the quality of decisions at the top of government. It affects defenders lower down through two routes. One is better attribution and policy response, which may deter some adversaries over time. The other is better products shared down through operational centers. The second depends on those centers having the staff to translate classified assessments into shareable guidance, and on recipients having the staff to use it. CTIIC’s design relied on both.
How intelligence products reach the field
Classified assessments reach defenders only after someone rewrites them. The usual vehicles in 2015 were joint DHS and FBI products, unclassified summaries with technical indicators attached, and briefings for cleared industry representatives. Producing a releasable version means separating what the adversary did, which defenders need, from how the government learned it, which must be protected. Analysts who can do that well understand both the intelligence and the networks of the people reading it.
Distribution rules added structure. The Traffic Light Protocol, widely used by sharing groups, marked how far a report could travel, from named recipients only to public release. Each marking decision is a judgment, and the people making it set the reach of every product. A center that produced excellent assessments for policymakers did nothing for field defenders unless that rewriting and marking work was staffed somewhere downstream.
The brokers: the NCCIC and a new class of sharing groups
The executive order addressed the middle of the chain: the organizations that pass information between government and everyone else. It made the NCCIC the government’s hub for agreements with ISAOs, and it set out to multiply the ISAOs themselves (Executive Order 13691).
Why the existing sharing centers were not enough
Sector-based Information Sharing and Analysis Centers had existed since the late 1990s. Financial services, health care, energy, state and local government and other sectors had them. They worked best for large members with their own security teams, which could contribute information and use what they received. DHS describes ISACs as one sector-based type of the broader ISAO category (CISA ISAO FAQ).
Many organizations fell outside the sector model or could not use it well. Law firms, accounting practices, small manufacturers and regional groupings of businesses had no ISAC built for them. The order’s answer was to let ISAOs form around any shared interest, including region. A group of companies in one metropolitan area, or a set of firms in one profession, could then organize their own sharing.
What a broker actually does
Brokering threat information is labor-intensive. A broker receives reports from members and government, removes identifying details, checks quality, adds context and redistributes. It runs a portal or feed, maintains trust among members and decides what can be shared with whom. It needs staff who understand both the threats and the members’ systems.
Large ISACs funded this work through membership fees, often tiered by member size. A new ISAO for small firms faced a harder problem. Its members could pay little, yet needed more help than large ones. The order created a framework for such groups to exist and a standards process to guide them. It did not fund their staff.
How brokers paid for themselves
The funding models available to an ISAO each carried trade-offs. Membership fees tied the group to what members could pay, which was least for those who needed the most help. Grants funded start-up costs but rarely recurring analyst salaries. Sponsorship by a few large members could subsidize smaller ones but gave the sponsors outsized influence. Groups run by security vendors could offer services cheaply but mixed sharing with sales, which made some members wary of contributing.
The executive order took no position among these models, which was reasonable for a voluntary framework. The practical result was that ISAOs formed where there was money: in sectors and regions with larger organizations willing to pay. Communities of small public bodies, which had the greatest need, were the least able to support a broker of their own.
Trust as an operating cost
Sharing depends on members believing that what they report will be handled carefully and will not be used against them. Building that trust takes staff time: vetting members, running meetings, enforcing handling rules and responding when a member’s report leaks. Mature ISACs had spent years on it. A new ISAO had to start from nothing, and its first analysts spent much of their time on relationships rather than analysis. That work was invisible in any count of indicators shared, but without it members contributed little.
The standards organization’s task
The Standards Organization was meant to make starting an ISAO easier by setting out how to do it: membership agreements, operating procedures, technical formats and privacy protections (Executive Order 13691). Standards reduce the design work for each new group. They cannot supply the analysts who run it. A small ISAO that adopted every recommended practice would still need someone to read incoming reports and decide what mattered to its members.
Clearances as a workforce issue
The order’s changes to the National Industrial Security Program let DHS decide whether ISAO members could receive clearances and access classified information. This addressed a real complaint: much of the most useful government threat information was classified, and private-sector defenders could not see it.
Clearances, though, are attached to individuals, and they take time to process. A cleared analyst needs a secure place to read classified material and must then work out what can be passed on without breaching classification. For a large bank, sponsoring a few cleared staff was feasible. For a small ISAO or a county government, the process consumed staff time that was already short, and the people cleared tended to be the ones most likely to be hired away.
Machine-speed sharing and its human cost
By early 2015 DHS was moving toward automated exchange of indicators using structured formats, so that sharing could happen in seconds rather than days. Automation reduces the labor of moving information. It increases the volume that arrives at each recipient. Without filtering and context, a feed of thousands of indicators is a burden on the team that has to decide which ones to block. Automation helps the broker more than it helps an understaffed consumer.
The NCCIC as a broker with its own limits
The order made the NCCIC the government’s counterpart for ISAO agreements (Executive Order 13691). The NCCIC already ran a round-the-clock watch, sent analysts to help victims of major incidents and published alerts. Each new agreement and each new ISAO added relationships to manage, reports to receive and requests for help to answer. Its staff were drawn from the same scarce pool as everyone else’s. A hub that grows its connections faster than its analysts becomes a switchboard: it can pass messages along but has less time to add context to them. That was a real risk for a center asked to serve every sector and every new sharing group at once.
The consumers: where indicators become protection
The end of the chain is the organization that receives a warning and has to act. Its staff must judge whether the threat applies, check whether it is already present, update controls and sometimes hunt through logs. This is the link that determines whether sharing protects anyone. The week’s announcements said little about it.
The work an indicator creates
Consider what a single shared indicator requires. An analyst reads it and decides whether it is relevant to the organization’s systems. An engineer adds it to the right blocking or detection tools without breaking legitimate traffic. Someone searches historical logs to see whether the indicator has already appeared. If it has, an investigation begins. Each step requires a person with skills and time, and access to logs that may not be kept.
Large organizations had security operations centers that performed these steps routinely. Most smaller organizations did not. Many relied on a general IT staff who also ran help desks, networks and applications. For them, a well-produced warning could arrive and go unread, or be read and not acted on because no one had the time or access to act.
Government consumers in particular
State and local governments were among the least able consumers. Counties, cities, school districts and utilities run critical services, hold sensitive data on residents and are frequent targets. Many had one or two people responsible for all of IT. The Multi-State Information Sharing and Analysis Center, supported by DHS and run by the nonprofit Center for Internet Security, served this community, and its services depended on federal funding (Center for Internet Security).
Federal agencies varied as well. Large departments had mature operations centers. Small agencies and commissions often had minimal security staff and relied on shared services or contractors. The NCCIC was meant to help them, but its own analysts were finite.
Why the consumer link was hard to fund
Consumer-side capacity is spread across thousands of organizations, each with its own budget. No single federal action can staff it. Grants can help, but they tend to fund equipment and projects more readily than recurring salaries. Shared services, where one team serves many organizations, are the main way to concentrate scarce skills. The executive order’s regional ISAOs were a step in that direction, but only if members were willing to pay for shared analysts.
What a capable consumer looked like
A capable consumer did not need a large team. It needed a named owner for threat intelligence, even part-time, who subscribed to the right feeds and filtered them. It needed central logs kept long enough to search. It needed blocking and detection tools that could accept indicators quickly, ideally automatically. It needed a relationship with someone who could help when an indicator matched, whether an ISAC, a state agency or a contracted provider. Each of these was within reach of a mid-sized organization that decided to prioritize them. Smaller organizations usually needed to buy them as a service.
Logs that exist when they are needed
Many shared warnings ask the recipient to check whether an indicator appeared in the past weeks or months. That check depends on logs being collected centrally and kept long enough to search. In 2015 many small organizations kept firewall and server logs only on the devices that produced them, for short periods, and had no tool to search across them. When a relevant warning arrived, the answer to whether they had been hit was often unknowable. Central logging is a technology purchase, but its value depends on someone who knows how to query it, which makes it a workforce decision as well.
The skills behind the work
Acting on shared intelligence draws on several distinct skills. An intelligence analyst judges relevance and context. A detection engineer turns indicators and behaviors into rules for security tools. An incident responder investigates when something matches. A systems administrator applies patches and configuration changes. In a large organization these are separate jobs. In a small one they are often the same person, who also answers the help desk.
Training could help a generalist do the first steps: subscribing to relevant feeds, reading alerts against an inventory of the organization’s systems and knowing when to call for help. It could not create the time to do it. The workforce constraint was hours as much as skills. A small agency that assigned threat intelligence to someone without removing other duties had assigned it to no one.
Measures that show whether sharing works
Few organizations in 2015 measured the consumer link. The useful measures were simple. One is the time from receiving a warning to deploying any resulting block or detection. Another is the share of relevant warnings that led to action. A third is how often shared indicators matched activity in the organization’s own logs, and a fourth is how often the organization reported something back. These measures show whether the people at the end of the chain have the time and tools to do the work. Counts of feeds subscribed or indicators received show only volume.
The return flow
Sharing is meant to run both ways. A consumer that finds a shared indicator in its logs, or spots a new attack, has information others need. Reporting it back requires the same scarce staff, plus approval to share. Organizations without spare capacity tended to consume without contributing. Brokers then had less to redistribute, and the network’s value depended on a small number of large contributors. Understaffing at the consumer end therefore weakened the whole chain as well as the consumer.
Shared services as the realistic answer
For organizations too small to staff the consumer link alone, the workable model was shared capacity. A state security office could monitor county networks under agreement. A regional consortium of school districts could fund one analyst for all of them. A sector group could offer a managed service tier to members without their own security operations. Each concentrated scarce skills where they served many organizations at once.
Shared services had their own requirements. They needed agreements on data access and authority to act, a way to route alerts to someone on site and stable funding beyond a single grant cycle. They also needed each member to keep a local owner, because the shared team could recommend action but rarely carry it out on systems it did not run. Where those conditions held, small organizations received the benefit of sharing that the February 2015 structures promised.
The gatekeepers: privacy, legal and classification review
Information sharing is constrained by rules about what may be shared. The executive order recognized this. It required agencies to build privacy and civil liberties protections into their sharing, based on the Fair Information Practice Principles, and to report assessments to DHS’s privacy and civil liberties offices (Executive Order 13691). CTIIC’s handling of information was likewise bound by existing law and policy on privacy and civil liberties (White House fact sheet).
Why review requires people
Threat reports can contain personal information: email addresses of phishing victims, names in malicious documents, logs that show employees’ activity. Before sharing, someone must remove what is not needed. Companies also worried about liability, antitrust and the confidentiality of what they disclosed. Their lawyers had to approve sharing arrangements, and in the absence of legislation, many approached them cautiously.
In government, classification review added another step. Analysts had to produce versions of classified reports that could be released to uncleared recipients. Doing this quickly and well requires experienced people who understand both the intelligence and the needs of recipients.
Where review becomes a bottleneck
Each gatekeeping step protects a real interest. Each also adds time. A warning that takes days to clear legal and classification review may arrive after an attack has moved on. Organizations that did not invest in their gatekeeping functions, by writing pre-approved sharing rules, standard redaction procedures and clear authority to share, found that review slowed everything else.
What a sound gatekeeping process looked like
The practical approach was to decide in advance. An organization could define classes of information that could be shared without case-by-case review, such as indicators stripped of personal information, and those that needed approval. It could give its security team delegated authority within those rules. Its privacy office could review the process once rather than each report. Agencies could do the same with standing release guidance for common types of classified information. This converted gatekeeping from a queue into a policy, which needed fewer people per report.
The privacy assessments the order required
The order’s privacy section had a workforce dimension too. Senior agency privacy officials had to assess their agencies’ sharing activities and send those assessments to DHS (Executive Order 13691). Agency privacy offices were small, and cyber sharing was one of many programs they reviewed. Reviewing sharing arrangements well requires privacy staff who understand technical data flows. Few agencies had them in 2015.
Member agreements and handling rules
For ISAOs, gatekeeping was written into membership agreements. These set out what members could share, how reports would be handled, who could see them and what members could do with what they received. The Standards Organization was charged with recommending model approaches to such contracts and procedures (Executive Order 13691). Good agreements reduced the review burden on each member, because their lawyers could approve the framework once. Poor or ambiguous ones pushed review back onto every report, and the members with the smallest legal teams shared least.
Public agencies faced added constraints. State public records laws could make information shared with an agency subject to disclosure requests unless an exemption applied. Agencies needed counsel to confirm how their state’s law treated threat reports before they could promise confidentiality to partners. Organizations should check these questions with counsel rather than assume federal protections carried over to state law.
Who can decide to share during an incident
Gatekeeping matters most during an incident, when sharing quickly could warn others and when lawyers, communications staff and executives are all cautious. An organization that has not decided in advance who may share what, and with whom, will usually share late or not at all. A short, pre-approved incident sharing rule changes that. It can name the role that may release technical indicators to the sector group or the NCCIC, and the information that requires executive approval. It also shortens the gatekeeper’s queue at the moment the queue matters most.
How the gap differed by type of organization
The same chain looked very different depending on who sat at the consuming end.
Federal departments
Large departments had operations centers, cleared staff and direct relationships with the NCCIC. For them, CTIIC and the order improved coordination with peers and the quality of incoming information. Their constraint was competition for skilled staff and the pace of hiring. Smaller agencies relied on shared services and DHS programs and gained less directly.
State governments
States had a chief information security officer and usually a central security team. Many also operated fusion centers that combined law enforcement and homeland security intelligence. They were plausible hubs for regional sharing with their counties and cities. Their limits were budget, staff and the reach of their authority over independently elected local governments.
Public utilities
Municipal water systems, public power utilities and transit agencies ran operational technology as well as office networks. Shared warnings about control systems required specialized knowledge to apply. Large investor-owned utilities had dedicated teams for this. Small public utilities often had a single engineer who knew the control systems and had no security training, and an IT department that understood security but not the plant. Acting on a control system warning required both, and few small utilities had a routine for bringing them together.
Counties, cities and school districts
Local governments faced the largest gap between threat and capacity. They received warnings from the Multi-State ISAC, from states and sometimes from the FBI. Many had no one whose job was to act on them. Their realistic path was shared capability: state-provided services, regional consortia or contracted monitoring. The February 2015 structures offered them more information, not more help.
Hospitals and health systems
Large health systems had security teams and participated in the health care sector’s sharing center. Community and rural hospitals often had small IT departments and little security staff. The Anthem breach had shown that health care was a target. A small hospital receiving an alert had the same problem as a small county: someone had to read it and act.
Tribal and territorial governments
Tribal nations and territories had sovereign responsibilities for services similar to those of states and counties, often with smaller budgets and more remote infrastructure. They were eligible for state and local sharing services, but their relationships with federal and state programs varied. For many, the realistic source of help was a shared provider that served them alongside other public bodies, which again made the funding of shared services the deciding factor.
Banks and credit unions
Large banks had mature intelligence teams and were the strongest members of their sector’s sharing center. Community banks and credit unions depended more on their core processors, managed service providers and the sector center’s summarized products. Their examiners expected them to have a way to receive and act on threat information, which made the consumer link a supervisory topic for them in a way it was not for most local governments.
What the comparison shows
The organizations best placed to benefit from better sharing were those that already had analysts. The structures announced in February 2015 increased the flow of information toward organizations that could absorb it, and toward those that could not. For the second group, the problem was staffing, and only shared services or direct funding could solve it.
The objection: sharing structures had to come first
The strongest objection is that the order was the right first step. Without a framework for sharing, there would be nothing for consumer-side analysts to work with. Building the pipes before the endpoints is a normal sequence, and the federal government could not staff thousands of independent organizations in any case.
Where the objection is right
The government’s levers were limited. It could reorganize its own intelligence work, encourage private organizations and use DHS’s authorities. It could not hire analysts for every county or hospital. The order did what an executive order could do without new money or legislation. Better integrated intelligence and clearer sharing rules were preconditions for any consumer-side improvement.
Where it falls short
The objection assumes that consumer capacity would follow once sharing improved. In fact, it lagged for years. More information arrived at organizations that could not use it. Shared services, the most promising answer, depended on stable funding that the federal government did not commit in 2015. The design treated sharing as a supply problem when the larger shortfall was on the demand side.
A different emphasis was available. Policy could have treated shared analyst capacity for small public and nonprofit organizations as part of the sharing architecture, rather than as a separate grant issue. Regional ISAOs serving local governments, with stable funding for their staff, would have connected the middle and end of the chain. That model appeared later, in parts of the country, mostly through state initiatives.
A second objection: the market would supply the analysts
Another objection holds that consumer capacity was a market problem. Managed security service providers sold monitoring and threat intelligence to organizations of every size, and small agencies could buy what they could not build. On this view, government had no reason to staff the consumer end.
The market did grow, and many organizations bought monitoring services. Purchasing still required budget, procurement capacity and someone to manage the provider. A provider monitoring a county’s network still needed someone at the county with the authority to block traffic, patch systems and decide when to escalate. Contracted services moved some work outside the organization. They did not remove the need for a capable owner inside it, and they were least affordable for the organizations with the thinnest budgets.
Questions a program director could ask now
- Who in our organization reads incoming threat warnings, and how much of their week is set aside for it?
- Which warnings did we act on in the last quarter, and how long did each take from receipt to action?
- Can our blocking and detection tools accept shared indicators automatically?
- Do we have pre-approved rules for what we may share back, or does every report wait for legal review?
- If our main sharing provider lost its funding, who would replace the analysis it provides?
With hindsight
The structures announced in February 2015 mostly came into being. In September 2015 DHS selected a team led by the University of Texas at San Antonio as the ISAO Standards Organization. Its partners were the Logistics Management Institute and the Retail Cyber Intelligence Sharing Center, and the award ran for up to five years (CISA ISAO FAQ). The Standards Organization published voluntary guidance on forming and running ISAOs, and many new groups formed around regions, professions and technologies.
Congress supplied the legal protections the January proposals had sought. The Cybersecurity Information Sharing Act of 2015, enacted in December 2015, gave companies liability and other protections for sharing cyber threat indicators under its terms. DHS built an automated indicator sharing service to carry those indicators. In 2018 Congress reorganized DHS’s cyber functions into a new Cybersecurity and Infrastructure Security Agency, which absorbed the NCCIC’s role.
The sharing law was enacted with a ten-year sunset. It lapsed on September 30, 2025. The funding bill that ended that autumn’s government shutdown, enacted November 12, 2025, restored it through January 30, 2026 (Covington, Inside Privacy). It has since continued on short extensions attached to funding bills. The Congressional Research Service’s most recent update reports the provisions extended to December 11, 2026 under a continuing appropriations act for fiscal 2027 (Congressional Research Service). Long-term reauthorization was still unsettled as this was written. Companies and their counsel had to track a moving deadline for protections they had relied on for a decade.
CTIIC operated for a decade inside the Office of the Director of National Intelligence. In August 2025 the Director of National Intelligence announced a reorganization called ODNI 2.0. It described CTIIC as redundant and said its work would move to the National Intelligence Council, according to the plan’s fact sheet and trade press reporting. The policymaker-facing integration role the center was created for became one task among many for other offices.
The consumer end of the chain had the hardest decade, and the clearest evidence came from local government. DHS announced on September 29, 2025 that it would not continue funding the Multi-State ISAC’s services for state, local, tribal and territorial governments. Federal support ended after September 30, 2025, and the Center for Internet Security moved the service to a fee-based membership model on October 1 (Center for Internet Security). Trade press reported that thousands of local members did not take up paid membership, while a number of states bought statewide memberships to cover their local governments. The organizations least able to staff their own analysis lost a shared provider at the same time as the legal framework for sharing became uncertain.
What held up from February 2015 was the brokering model. ISACs and ISAOs became a normal part of how organizations received threat information, and automated indicator exchange became routine. What did not hold up was the assumption that better flows of information would, by themselves, produce better-defended recipients. The constraint the week’s announcements left untouched, analyst capacity at small public and private organizations, remained the weakest link. In some places it grew weaker when federal support for shared services ended.
By late 2026, a program director at a county or small agency should assume two things. Threat information will keep arriving from many sources, and the work of turning it into action is the agency’s own. Shared services remain the most efficient way to buy that capacity. Their funding now has to be planned locally rather than assumed from Washington.
Spotted an error? Tell us and we will check it and correct the article where needed.