The November 7, 2025 forum: edibles pass, the hard tests slip to 2026, and the trust problem gets said out loud

Colorado's Science & Policy Forum met November 7, 2025, the last meeting before the January 5 testing rules took effect. We transcribed the Division's published recording (about 2 hours 4 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. Two things happened: the surveillance program reported its first real results and its first real delay, and the forum's topic-setting discussion for 2026 turned into the frankest airing of the lab-trust problem in the archive.

Notes from the editor

The broadside in this meeting ("it has never been cheaper to get the results that I want and more expensive to get an accurate understanding of what is in my inventory") was one angry stakeholder's claim in November. Watch what the record did with it. By June the Division's own science director was naming lab-shopping on the record; by July the state's data showed lab-passed samples surviving a state retest at roughly a coin flip. His specific claims stayed contested, and this site adopts none of them. But the state's own data later walked most of the way toward his diagnosis, and that's where the credit belongs: to the data, not the volume. We can also now say the machinery his sentence describes is real and documented by the regulator itself. In January 2026 the Division's own bulletin named licensees using R&D testing (the consequence-free pre-screen a lab will run without reporting a failure) to circumvent required testing. That is, mechanically, what "the results that I want" being cheap looks like. When angry testimony and the agency's own documents converge, you don't have to take anyone's word for it.

One more piece of context the public enforcement record supplies for this meeting's unanswered lab-accountability question: the year's one potency-driven laboratory action had already happened: a Colorado lab appears in the MED's 2025 final actions after investigators concluded, per trade-press reporting we have not independently verified in full, that it was inflating potency data, with its certification lost. The lab representative's question here (whether revocation causes and affected samples could be made public) was asked into exactly that silence, and it still stands.

The cleanest promise of the year was also made here: CDPHE committed to making its surveillance testing methods public. And, credit where due, it has been substantially kept: as of August 2026, CDPHE has publicly posted five of its reference-lab methods (potency, heavy metals, salmonella/STEC, total yeast and mold, and Aspergillus, the surveillance panel), redacted only for vendor names. The pesticide and residual-solvent methods, still in validation, have yet to appear; we're watching for those two.

Since this meeting

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

Resolved. The early microbial pass held up: the full June results confirmed every safety test passed: microbials and metals, all 45 samples. The 2026 agenda auctioned here was settled at the January meeting, and interlab consistency became the year's organizing topic exactly as flagged.

Still in progress. The methods commitment was substantially delivered: technical method-review sessions began in February and continued at a one-to-two-month cadence, and by August 2026 CDPHE had publicly posted five reference-lab SOPs covering the surveillance panel. The pesticide and solvent methods, still in validation, have yet to appear. Method performance requirements reached the full forum in August.

Put off or neglected. The Q1 2026 target for flower and concentrate surveillance was missed by two seasons: April moved it to fall 2026. The paired-sample comparison a producer offered (his own COA against the state's result on the same batch) was taken under advisement and never resurfaced in the recordings we transcribed. Neither did the lab-accountability question: whether certification revocations and their affected samples could be made public.

What was brought up

Phase one, executed [0:05:06]. The October collection went almost exactly to plan: 45 edible samples over six days from 16 stores in seven cities, representing 19 manufacturers. Store staff had read the bulletin, knew the program, and, per the Division, were often enthusiastic about being sampled. Early results from the state lab: every edible microbial test passed, with potency, cannabinoid, and metals testing finishing within the week [0:15:14].

The delay [0:10:10]. The health department's chief medical officer announced that pesticide and residual-solvent methods (the tests that matter most for flower and concentrates) needed more validation time: pesticide testing alone requires validating more than 100 compounds. Flower and concentrate surveillance slipped to Q1 2026, and sample collection paused rather than collect what couldn't yet be fully tested. The lab's new environmental-testing deputy director was equally direct about the cause (infrastructure problems, including HVAC issues, since resolved) and the framing: "the pause is about quality, not capability" [0:15:14]. Asked directly whether money was the constraint, the chief medical officer said no [0:20:18].

The trust problem, said plainly [0:45:34]. An infused-products manufacturer delivered the year's bluntest statement of the interlab crisis: no shared standards or methods, certified reference materials for THC varying widely, isolate potency results spanning 85–106% on the same material, and a regulatory structure that, in his words, rewards following a process rather than being accurate: "it has never been cheaper to get the results that I want and more expensive to get an accurate understanding of what is in my inventory" [1:55:48]. He also disclosed active litigation against the department over adverse-event records, and made claims about the department's past conduct that the co-chair disputed on the record in the moment [0:50:36]. The chair repeatedly steered the exchange back to specifics; the state lab's leadership answered with commitments rather than counterattack. Most concretely: CDPHE will make its surveillance methods public [0:55:37].

The 2026 agenda auction [1:10:39]. The meeting's second half solicited and argued next year's topics: interlab consistency (with a lab representative's counterpoint that method standardization is "a house on an unstable foundation" without randomized, non-self-collected sampling first [1:25:42]); indicators of synthetic cannabinoid production (supported, with the caution that mandatory tests burden honest operators while cheats cheat; surveillance-style testing catches more [1:40:46]); the ASTM universal-symbol proposal (argued down by multiple operators as policy-not-science, costly, and unproven against Colorado's existing symbol [1:30:43, 1:45:47]); metrc data quality, harvest-batch definitions, vape expiration dates, and a deregulation-and-efficiency review for a struggling industry [1:35:45].

What was nailed down

• All phase-one edible microbial results passed; remaining edible tests were days from completion, with an edibles-specific preliminary report under consideration rather than waiting for the whole program [0:15:14, 0:25:19].
• Flower and concentrate surveillance moved to Q1 2026, collection paused in the meantime, a new industry bulletin promised [0:10:10].
• CDPHE committed to making its testing methods public [0:55:37], a direct answer to the transparency thread that began in September.
• Individual passing results won't be routinely shared with the sampled businesses; reporting stays aggregate except where failures trigger action [0:20:18].
• The rigor question got a flat answer: asked whether the state lab's validation standards were consistent with or higher than what licensed labs face, the answer was "consistent" [0:55:37].

What was left open

• Everything about interlab consistency: whether the fix starts with shared methods, minimum performance requirements, sampling reform, or a state-declared benchmark; whether lab methods are intellectual property or "public goods"; and whether the state should simply convene the labs and pick a standard versus the state lab's caution that mandated methods have failed badly in other states [1:15:40, 1:20:41, 1:25:42].
• Self-collected samples: named by a lab representative as the foundation problem: labs can verify only that they received the regulatory minimum, not that it represents the batch [1:25:42]. This becomes the third-party-collection debate of 2026.
• Lab accountability transparency: a lab representative noted a certification revocation at a prominent lab had produced only one product recall, and asked whether revocation causes and affected samples could be made public. No commitment was made [1:55:48].
• Whether a paired-sample comparison (a producer's own full-panel COA against the state lab's surveillance result on the same batch) could be arranged. It was taken under advisement [0:20:18].
• The whole 2026 agenda: settled two months later at the January 9 meeting.

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.