Does poor sleep actually tank your testosterone during a cut?

What did @daviddemesquita actually say?

The creator claims that DSIP (delta sleep-inducing peptide) is a reliable sleep aid that works for "98% of people" he's used it with, that a personal sweet spot of 300-400 micrograms per day is broadly applicable, and that taking slightly too much (100 micrograms over his threshold) caused him to wake up actively dreaming, a phenomenon sometimes associated with REM sleep disruption. He also claims it can sustain sleep "10 plus hours even while blasting tren," referencing trenbolone, a powerful anabolic steroid.

The self-deprecating framing, "I am losing credibility," doesn't soften the fact that he's offering specific dosing guidance based entirely on personal experimentation. That matters, because DSIP has a thin clinical research record and essentially no regulated medical use in humans in the United States.

Does the science back this up?

Barely, and not in the way the creator implies. DSIP was first isolated in 1974 and generated real interest through the 1980s, but that interest largely faded because human trials produced inconsistent, context-dependent results.

The original work by Schoenenberger and colleagues (1977, Pflügers Archiv) described DSIP as a neuromodulator that could alter sleep architecture in rabbits. Subsequent human studies, including a small trial by Monnier et al. (1979, Pharmacology Biochemistry and Behavior), suggested it might reduce sleep latency and increase slow-wave sleep under specific conditions. However, a later review by Graf and Kastin (1984, Neuroscience and Biobehavioral Reviews) noted that results were highly variable and dependent on baseline sleep quality, peptide stability, and route of administration. DSIP degrades rapidly in plasma, which makes subcutaneous injection an unreliable delivery method for consistent CNS effects. The creator's anecdote about sleeping 10.5 hours is plausible in isolation, but attributing it specifically to DSIP's REM-forcing properties goes beyond what the literature supports.

What did they get wrong (or right)?

He got the basic concept half-right. DSIP does appear to influence sleep architecture, particularly slow-wave sleep, in some research contexts. His observation that taking more didn't just deepen sleep but created something closer to a dissociative, dreaming-while-awake experience is actually worth noting. That kind of experience is loosely consistent with REM intrusion or hypnopompic hallucination, not just "forced REM" as he frames it.

What he got wrong is more significant. First, the "98% success rate" is a fabricated statistic. He has no controls, no comparative group, and no way to isolate DSIP's effect from placebo, lifestyle changes, or other compounds in his clients' stacks. Second, presenting 300-400 micrograms as a general population sweet spot is irresponsible. There is no peer-reviewed human dosing protocol for DSIP. Third, using DSIP while also using trenbolone, a compound known to cause severe sleep disruption and androgenic side effects, means any reported sleep benefit is confounded to the point of being uninterpretable. Attributing sleep improvement to DSIP in that context is guesswork dressed as insight.

What should you actually know?

DSIP is not approved by the FDA for any therapeutic use. It is sold in research markets and compounded peptide suppliers, which means purity, concentration accuracy, and sterility are not guaranteed. The clinical history is real but limited, and most of the promising early data never translated into human clinical trials robust enough to establish safety or dosing standards.

If you're dealing with sleep issues related to hormonal therapy or bodybuilding compounds, that is a legitimate clinical problem worth addressing. There are evidence-based options including cognitive behavioral therapy for insomnia (CBT-I), which multiple meta-analyses, including Trauer et al. (2015, Annals of Internal Medicine) show outperforms sleep medications in long-term outcomes. Melatonin, magnesium glycinate, and addressing the underlying hormonal disruption with a licensed provider are also grounded options. Injecting an unregulated peptide based on a TikTok creator's personal dose experimentation is not a risk-neutral decision, regardless of how honest he sounds about it.