The January 9 forum: a year of testing fights gets scheduled, an edibles report nears, and a fan explodes at the state lab

Colorado's Science & Policy Forum held its 2026 kickoff on Friday, January 9, four days after the state's revised testing rules took effect. We transcribed the Division's published recording (about 1 hour 56 minutes) with a speech-recognition model and drafted this summary from that transcript; our editor has reviewed and approved this page, and the recording remains the source of record. The meeting was a planning session, and that's what makes it worth reading: nearly every testing argument Colorado would have in 2026 was scheduled here, in advance, on the record.

Notes from the editor

Read this one as the year's table of contents: nearly everything scheduled here actually ran, which is to the forum's credit. Then notice what was never scheduled: enforcement. The year's debates kept crashing into it: February's claim that half of recent enforcement actions involved sample adulteration, March's California lesson that enforcement rather than collector identity is what makes sampling honest, June's demand for minimum fines and license revocations. And the agenda never gave it a meeting of its own.

We've since checked the other half of that story in the state's own records. Our sweep of the complete 2026 public enforcement record (every final administrative action, advisory, and bulletin posted as of mid-August) found a busy enforcement year, running almost entirely on contaminants, with the label-truth side essentially empty: one potency case in thirty-nine actions, opened back in 2024. And the Division showed in January what its openness looks like when it's willing: two weeks after this meeting, its own bulletin publicly described Reduced Testing Allowance abuse and R&D-testing circumvention: licensee gaming, named in writing, on the contaminant side. No equivalent written statement about the label side appeared all year, even after the surveillance data landed; the closest the record comes is the science director naming lab-shopping aloud in June.

And the question a manufacturer raised in this meeting's first half hour (is the goal labs consistent with each other, or consistent with an anchor?) was still unanswered on August 7, when the gap between lab results and the state lab's drew "I don't think we have an answer to that." The year ran its whole agenda without settling what its headline word meant.

Since this meeting

Where the threads this meeting opened stand as of August 2026, from the recordings that followed.

Resolved. The calendar mostly executed: third-party sampling ran from February through June to a no-consensus close; the proficiency-testing deep-dive landed in April; the analyte process was built across the spring and its submission form went live in July; and the SOP technical breakouts happened, starting February 19.

Still in progress. Method performance requirements (the third consistency subtopic named here) only reached the full forum in August, feeding the fall rulemaking season.

Put off or neglected. The edibles report promised "before the February meeting" wasn't presented in full until June. And the written summary and report the program's founding bulletin committed to (roughly 30 and 90 days after completion) had, as of mid-August 2026, never appeared on the open web at all: the findings live in meeting recordings and a sign-up-gated folder. The exploded fan proved prophetic: repairs begat repairs, surveillance resumption slid from Q1 to fall 2026, and the case for actually funding the facility is now its own article. And the definitional question was scheduled, argued all year, and never resolved.

What was brought up

The year's two umbrella topics [0:10:12]. The co-chairs (the Marijuana Enforcement Division's deputy senior director and the state public health laboratory's deputy director of policy and regulatory affairs) proposed organizing all of 2026 under two headings: interlab consistency (why different labs get different numbers from the same product) and exploration of additional analytes (whether Colorado should test for more things than it does).

Interlab consistency came with three scheduled subtopics: a February deep-dive into the proficiency-testing program (the periodic send-every-lab-a-known-sample exercise that checks whether labs get the right answer); at least two meetings on minimum method performance requirements (shared standards for detection limits, quality-control recovery ranges, and validation, so that different instruments must still perform equivalently) [0:15:14]; and a June–July examination of required third-party sample collection (taking the sample-picking step out of licensees' own hands) [0:20:20].

A process for new analytes, not just new analytes [1:00:53]. On the second umbrella, the Division proposed building a standardized evaluation framework (required fields covering public-health need, prevalence data, sampling impacts, equipment and method burdens, and cost) that any proposed new test (chromium, mycotoxins, and synthetic THC tests were all suggested last year and not adopted) would have to pass through. A manufacturer's representative called the idea the meeting's most valuable potential work product: define the process once, because "this question is going to come up again — if it's chromium this year, it's something else next year" [1:11:00].

The SOP review breakout [1:16:00]. The state lab offered to share its own reference-laboratory testing procedures with a small technical working group (likely application-gated, hybrid or in-person at CDPHE, offset between monthly forums), starting with the methods used in the surveillance program (potency, yeast and mold, salmonella and STEC). Lab and industry participants pushed to prioritize test-method SOPs and validation reports, with one lab representative specifically requesting the pesticides method be among the first reviewed [1:26:14].

The surveillance program update [1:31:16]. Phase one collected 45 edible samples across 16 stores in seven cities (Denver, Broomfield, Aurora, Lakewood, Frisco, Garden City, Colorado Springs) from 19 manufacturers in October and early November 2025. Testing is complete, and a report of the results was drafted and in internal and MED review, promised to the group before the February meeting.

And the fan [1:36:28]. A rooftop air-handling unit at the 30-year-old state laboratory building suffered an exploded fan; inspection found failing bearings in three of the four other units, which were shut down before they could fail the same way. The destroyed unit was estimated at a five-week repair (into early March), and the lab flagged, transparently, that the resumption of flower and concentrate surveillance testing targeted for Q1 could slip as a result.

What was nailed down

• The 2026 calendar: February for proficiency testing, roughly March–April for method performance requirements, June–July for third-party sample collection, with July as the wrap-up month because the forum's recommendations feed the Division's rulemaking cycle [0:20:20].
• The edibles surveillance report exists: drafted, in review, and committed for release ahead of the February meeting [1:36:28].
• The SOP breakout will happen, structured as a small technical group rather than a full-forum session, with CDPHE planning the agenda and logistics [1:31:16].
• Why flower surveillance paused: sample collection was suspended in November so resources wouldn't be spent collecting samples the reference lab couldn't yet test for pesticides (flower) or solvents (concentrates); resumption tied to those methods coming online [1:31:16].

What was left open

• What "interlab consistency" even means. An infused-products manufacturer argued the term is "a bit of a myth" and that the real question is whether third-party labs match the state reference lab, "the North Star," and that without a shared reference point "we can't do precision without accuracy" [0:25:23, 0:30:26]. A trade association representative sharpened it to the underlying philosophical question: "is there a correct result, or are we going to continue... just measuring the methodology?" [0:40:31]. The forum scheduled the discussion; it did not resolve the definition.
• Whether proficiency testing should become continuous. Two stakeholders proposed folding lab oversight into the surveillance program itself: splitting shelf samples across the reference lab and multiple licensed labs as a running "ring test," on the argument that a scheduled test is like a restaurant inspection announced in advance [0:45:38, 0:50:42]. The state lab noted its potency proficiency program already runs twice a year with international-standard homogeneity verification, but acknowledged contaminant ring-testing faces a real barrier: someone has to introduce contaminants into samples, and "we're not really funded to do that work" [0:50:42].
• Third-party sample collection: the whole fight, previewed. One manufacturer called eliminating licensee self-collection "the most important" reform, citing documented episodes of sampling problems [0:30:26]. A multi-state operator counseled caution: for homogeneous manufactured products the improvement may be small against a "dramatic cost increase," and other states' third-party regimes have their own flaws worth studying first [1:11:00]. June and July will host that argument.
• The cost question, standing. A manufacturing operator asked the Division to weigh every change against "actual harm" to consumers and the costs passed to licensees: "it is very tough out there right now for businesses" [1:51:39]. Another stakeholder's list for the analyte framework included cost analysis and federal rescheduling harmonization [1:06:00].
• The reference lab's structural limits. CDPHE's lab cannot act as a third-party testing lab for producers: its federal controlled- substances registration bars receiving marijuana from non-registrants, and no workaround has been found [0:55:52].
• The surveillance resumption date: Q1 2026 was the stated target, with the HVAC failure named as the risk to it [1:41:31].

This summary was drafted by our AI desk from a machine transcript and has been reviewed and approved by our editor (2026-08-13). The Division's published recording is the source of record. Names are used only for officials acting in public capacity; other speakers are identified by role because machine transcription garbles proper names.