The 2023 statute and the state jail felony grade
Texas Penal Code § 21.165, enacted by HB 2700 / SB 1361 in the 88th Legislature and effective September 1, 2023, is a state jail felony — 180 days to 2 years in state jail and a fine up to $10,000 — criminalizing the knowing creation or electronic distribution of deepfake intimate imagery without effective consent and with intent to harm.
- Knowingly creates OR distributes — § 21.165(b)(1)
- The statute prohibits two distinct conduct prongs joined disjunctively. A defendant violates § 21.165 by either (1) knowingly CREATING the deepfake material or (2) knowingly DISTRIBUTING it by electronic means. The State need only prove one prong, not both. This means a defendant who creates the material but never distributes it can be charged on the creation prong alone; conversely, a defendant who receives synthetic material from a third party and then redistributes it can be charged on the distribution prong without any involvement in the creation. The disjunctive structure is critical for defense strategy — counsel must identify which prong the State has charged and develop a defense focused on negating that prong specifically.
- Intent to harm, harass, intimidate, or coerce — § 21.165(b)(1)
- The mens-rea element specific to § 21.165 requires that the defendant acted with the specific intent to harm, harass, intimidate, or coerce the depicted individual. This is higher than the general "knowingly" mens rea and is the most contested element in synthetic-media prosecutions. Artistic creation, satirical commentary, educational use, or generic pornographic-style content directed at no particular individual lacks the targeted-harm intent the statute requires. Counterman v. Colorado, 600 U.S. 66 (2023), recently held that "true threats" prosecutions require at least recklessness as to the threatening nature of the communication — that framework is likely to inform First Amendment analysis of intent-to-harass statutes including § 21.165.
- Visual material appearing to depict another in sexual conduct — § 21.165(b)(2)
- The result element — the visual material must appear to depict another individual engaging in sexual conduct. "Sexual conduct" tracks the § 43.25(a)(2) definition imported across chapter 21. Critically, the statute uses "appears to depict" — recognizing that synthetic media does not actually depict the named individual but rather creates a representation that appears to depict that person. This phrasing is the textual hook for deepfake technology specifically and distinguishes § 21.165 from § 21.16, which requires that the material actually depict the named individual.
- Without effective consent — § 21.165(b)(3)
- The non-consent element imports the general Penal Code definition of effective consent at § 1.07(11). Consent given to be photographed clothed, filmed in a non-sexual context, or to appear in mainstream media is NOT consent to synthetic sexual depiction. The element is structured similarly to consent in § 22.011 and § 21.16 — the State must prove the absence of consent specific to the conduct charged. Consent obtained by fraud (for example, falsely promising the material would not be shared) is not effective consent under § 1.07(11)(A).
- Penalty under § 12.35 — state jail felony range
- Conviction carries 180 days to 2 years in a state jail facility plus an optional fine up to $10,000. State jail felonies are served in state jail facilities (distinct from TDCJ-CID prisons) and do not generate good-conduct credit or parole eligibility in the conventional sense — state-jail time is served day-for-day, though some early release through Code Crim. Proc. art. 42A.554 mandatory supervision can apply. Critically, § 21.165 conviction does NOT trigger mandatory sex-offender registration under Code Crim. Proc. art. 62.001 unless the underlying conduct also violates § 22.011 or § 43.26 — a structural distinction from most other chapter 21 offenses that makes § 21.165 plea posture meaningfully different from § 21.11 or § 21.02 plea posture.
The 2023 enactment was Texas's direct legislative response to the proliferation of generative-AI tools capable of producing photorealistic synthetic intimate imagery of identifiable individuals. The 88th Legislature, observing the gap between existing § 21.16 (which requires real imagery) and the emerging harms of synthetic-media misuse, enacted HB 2700 / SB 1361 to create a parallel state-jail-felony framework specifically targeting deepfake content. Texas was among the early adopting states in this space — by mid-2024, more than half of U.S. states had enacted some form of nonconsensual-deepfake legislation, with the contours of each statute varying significantly. Section 21.165 is at the more cautious end of that legislative spectrum: it requires both specific intent to harm and lack of effective consent, structures that the legislature inherited from the post-Jones § 21.16 redraft and designed to maximize First Amendment durability.
The state-jail-felony grade is significant. A state-jail felony in Texas carries 180 days to 2 years in state jail (not TDCJ prison) plus a fine up to $10,000. State jail time is generally served day-for-day, though Code Crim. Proc. art. 42A.554 provides for mandatory supervision under certain conditions. Probation and deferred adjudication are generally available; the offense is not enumerated as a 3g aggravated offense under Code Crim. Proc. art. 42A.054. Where the case can resolve at deferred adjudication, the defendant may emerge without a final conviction on record — a meaningful collateral-consequences outcome given the deep social and professional impact of any sex-offense-related entry in a background check. The non-registration default under art. 62.001 is the most important structural feature of § 21.165 plea posture: most plea negotiations focus on resolving the case in a manner that preserves the non-registration outcome and avoids the parallel § 43.26 or § 22.011 exposure that would trigger lifetime registration.
Elements, intent, and the creation-versus-distribution prongs
Section 21.165 has two disjunctive conduct prongs — knowing creation OR knowing electronic distribution — joined to a single specific-intent element requiring intent to harm, harass, intimidate, or coerce. The defense must identify the charged prong and develop a defense specifically tailored to that prong's proof requirements.
The element structure of § 21.165 is more complex than the statute's short text suggests. The State must prove four distinct elements: (1) the defendant acted knowingly; (2) the defendant either CREATED or DISTRIBUTED the material (one or the other, joined disjunctively); (3) the material is visual material appearing to depict another individual engaging in sexual conduct; (4) the depicted individual did not give effective consent; and (5) the defendant acted with specific intent to harm, harass, intimidate, or coerce that individual. The mens-rea structure combines a general "knowingly" requirement (covering the conduct elements) with a specific-intent requirement (covering the targeting element) — this two-tier mens rea is unusual in the Texas Penal Code and tracks the structure of the stalking statute under § 42.072.
The creation prong is satisfied where the defendant used generative-AI tools (Stable Diffusion, MidJourney, DALL-E variants, specialized "nudify" applications, or custom-trained models) to produce the synthetic visual material. Knowledge of the depicted individual's identity is structurally required — the defendant must have intended to create material appearing to depict a SPECIFIC person, not generic synthetic imagery directed at no identifiable individual. The creation prong is technically easier to prove when forensic evidence demonstrates that the defendant's device or account was used to generate the specific material, and harder to prove when the material was obtained from third parties or shared communities where attribution is uncertain.
The distribution prong is satisfied where the defendant electronically transmitted, shared, posted, or uploaded the material to any platform or recipient. Electronic distribution under § 21.165 likely covers text messages, social media posts, file-sharing services, email, and direct messaging — the statute uses "by electronic means" broadly. A defendant who created material but never distributed it can be charged on the creation prong alone; a defendant who received material from a third party and redistributed it can be charged on the distribution prong without participating in creation. Counsel must therefore identify the State's actual theory — which prong is charged, what evidence supports it — and develop a defense focused on the contested prong specifically. A "no creation" defense (the material existed before the defendant's involvement) does not defeat a distribution prosecution; conversely, a "no distribution" defense (the material never left the defendant's control) does not defeat a creation prosecution.
The specific-intent element is where the bulk of § 21.165 litigation occurs. Intent to harm, harass, intimidate, or coerce is not satisfied by mere creation or distribution of sexually explicit material — the State must prove the defendant's conduct was directed at the depicted individual with one of those four targeted harmful purposes. Prior threatening communications, prior conflict with the depicted individual, contemporaneous statements explaining the defendant's motivation, distribution patterns directed at the depicted individual's known contacts or workplace, and evidence of attempted coercion (sending the material with demands or threats) are all common State evidence on this element. The defense argues lack of specific intent where the material was created for personal use only, distributed to communities unrelated to the depicted individual, posted with no targeting indicators, or created in an artistic or satirical context. Counterman v. Colorado, 600 U.S. 66 (2023), suggests that First Amendment analysis of intent-to-harass statutes will require some level of awareness of the threatening or harassing nature of the communication — a framework that may extend to § 21.165 prosecutions on direct appeal.
The "effective consent " element under § 1.07(11)
Section 21.165 requires the State prove the depicted individual did not give "effective consent" to the synthetic intimate imagery — a defined term at § 1.07(11) that imports specific limitations on consent obtained by fraud, threat, or from someone unable to make reasonable decisions.
"Effective consent" is a defined term across the Texas Penal Code under § 1.07(11). Consent is "effective" when given by an individual legally authorized to give it and not vitiated by one of four enumerated defects: (A) consent induced by force, threat, or fraud; (B) consent given by someone the actor knows is not legally authorized; (C) consent given by someone who by reason of youth, mental disease or defect, or intoxication is known by the actor to be unable to make reasonable decisions; or (D) consent given solely to detect the commission of an offense. The definition is global and applies wherever consent is an element of a Penal Code offense, including § 21.165.
The most important consent issue in § 21.165 cases is the SCOPE of any consent the depicted individual gave. Generic consent to be photographed clothed, filmed at a public event, or appear in social-media imagery is not consent to be synthetically depicted engaged in sexual conduct. The defense argument that "she posted bikini pictures online, so she consented to whatever happens to her image" misreads the consent element fundamentally — consent under § 1.07(11) is specific to the conduct that is the subject of the criminal element, not generic permission to use one's likeness for any purpose. Texas case law on consent in § 22.011 and § 21.16 routinely emphasizes specificity: consent to one act is not consent to a different act, and consent under one set of circumstances is not consent under different circumstances. Hughes v. State, 612 S.W.2d 581 (Tex. Crim. App. 1981), and the line of cases tracing consent scope in sexual-assault prosecutions inform how courts will likely read § 21.165 consent.
Consent induced by fraud is the second recurring issue. Where the defendant obtained an image of the depicted individual under false pretenses — for example, by representing that the image would be private, used only for a specific purpose, or kept off public platforms — and then used that image as a source for the deepfake, the consent element is potentially vitiated under § 1.07(11)(A). The fraud must be material to the consent given. Subtle deception about ultimate use is generally enough to invalidate consent for purposes of the criminal element; a defendant who said "I just want a regular photo of you" and then transformed that photo into a deepfake has obtained consent by fraud as to the photo and not at all as to the deepfake.
The intoxication and youth limitations under § 1.07(11)(C) create the practical bridge to § 43.26 (child pornography). If the depicted individual was under 18 at any time when the source material was created — even if the source material itself is innocuous — the case potentially crosses into § 43.26 territory and triggers mandatory registration. Practitioners must screen carefully: the deepfake intimate imagery of a 17-year-old created when the depicted individual was 17 generates § 43.26 exposure (and likely federal CSAM exposure under 18 U.S.C. § 2252A), not § 21.165 alone. Plea posture in any § 21.165 case where the depicted individual's age is contested or unclear should be paused until age can be conclusively established — a § 21.165 plea on what should have been a § 43.26 case is the kind of resolution federal prosecutors routinely reopen.
Defense strategies for § 21.165 prosecutions
The seven primary defense strategies for § 21.165 prosecutions are the First Amendment overbreadth challenge under Ex parte Jones and Hansen, insufficient specific intent, prong-specific negation, account-hijack defense, effective-consent argument, insufficient deepfake-identification proof, and federal-state charge-bargaining.
First Amendment overbreadth analysis is the threshold defense issue in any § 21.165 case. The statute is a content-based restriction on speech — it regulates the creation and distribution of visual material based on what the material depicts. The Texas Court of Criminal Appeals struck the prior version of § 21.16 in Ex parte Jones, 596 S.W.3d 257 (Tex. Crim. App. 2020), holding the statute facially overbroad because it punished a substantial amount of protected speech relative to its legitimate sweep. The legislature redrafted § 21.16 to add specific-intent and consent elements designed to narrow the statute and survive constitutional review. Section 21.165 imports the same narrowing structure but applies it to synthetic-media content. Whether § 21.165 will survive overbreadth analysis on direct appellate review has not yet been decided. United States v. Hansen, 599 U.S. 762 (2023), refined the overbreadth doctrine — requiring a "substantial" amount of protected speech be punished relative to the statute's "plainly legitimate sweep" — and the analysis in Hansen will be central to any future § 21.165 overbreadth challenge.
Insufficient specific intent is the most common factual defense. The State must prove the defendant acted with intent to harm, harass, intimidate, or coerce the depicted individual — not merely that the defendant created or distributed sexually explicit synthetic material. Where the material was created for private personal use only, distributed to communities not connected to the depicted individual, posted without any indicia of targeting (no captioning, no identification, no relationship-specific context), or created in an artistic or satirical mode, the specific-intent element is not satisfied. Defense counsel develops the absence-of-intent narrative through device-forensics review, distribution-pattern analysis (was the material sent specifically to people who know the victim?), and contextual evidence of the defendant's motivation at the time of creation or distribution. Where the State's intent evidence rests on inference from distribution patterns alone, that inference is contestable and frequently insufficient.
Prong-specific negation is the strategic move that distinguishes § 21.165 defense from § 21.16 defense. Because the statute joins creation and distribution disjunctively, the defense must identify which prong the State has charged and develop a defense specifically negating that prong. A "no creation" defense (the material existed before the defendant's involvement; the defendant never used generative-AI tools) does not defeat a distribution prosecution if the State can prove the defendant electronically distributed the pre-existing material. A "no distribution" defense (the material never left the defendant's sole control) does not defeat a creation prosecution if the State can prove the defendant created the material. Counsel must read the charging instrument carefully, request a bill of particulars where the prong is ambiguous, and develop the defense specifically for the prong charged. Mismatched defenses — defenses developed for the wrong prong — are among the most common preventable mistakes in this new area of practice.
Account-hijack and identity-mistake defenses arise frequently because the conduct occurs entirely in digital space. Where the State's evidence is that "the material was uploaded from this account" or "this device generated this image," the defense investigates whether the account or device was actually controlled by the defendant at the relevant time. Shared devices, compromised accounts, household network access by other individuals, and account theft are recurring real defenses. Forensic analysis of device usage patterns, login records, and timing relative to other known defendant activities can sometimes establish that someone other than the defendant was the operator at the moment of creation or distribution. The State bears the burden of proof beyond a reasonable doubt as to identity, and the digital-evidence chain is frequently weaker than it first appears.
Insufficient deepfake-identification proof is a technical defense unique to § 21.165. The statute requires that the visual material "appears to depict" a particular individual — meaning the State must prove the depicted individual is in fact the named complainant. In synthetic imagery, this is not as straightforward as it might seem. The defense engages forensic-imagery experts to analyze whether the depiction can be reliably attributed to the named individual, whether the source material used as input can be authenticated, and whether the synthesis quality is high enough to support an "appears to depict" finding under any reasonable identification standard. Generic synthetic pornographic content that vaguely resembles a real person but cannot be conclusively attributed to that person may not satisfy the appears-to-depict element. The technical proof in this area is still developing, and defense forensic experts can sometimes generate genuine doubt about identification that the State cannot overcome.
Federal-state charge-bargaining is the seventh strategic lever. Where federal jurisdiction is available — typically because distribution crossed state lines using interstate electronic communication — the case may be charged federally under 18 U.S.C. § 2261A (stalking), under the TAKE IT DOWN Act's 2025 federal criminal provisions, or under any future DEFIANCE Act enactment. Federal exposure must be calculated carefully: Federal Sentencing Guidelines exposure can be higher or lower than the state-jail-felony exposure under § 21.165, depending on prior history, offense conduct (including any minor-related conduct), and guideline application. Where federal exposure is meaningfully lower, defense counsel may seek to direct the case to federal venue (typically by alerting the AUSA office and the FBI of the conduct); where federal exposure is higher, defense counsel seeks to keep the case in state court and resolve it before federal interest develops.
The First Amendment overbreadth question
Section 21.165 has not yet received appellate overbreadth review. The Ex parte Jones framework that struck the prior § 21.16, the Hansen refinement of overbreadth analysis, and the First Amendment jurisprudence on content-based restrictions (Stevens, Brown v. EMA, Alvarez) all bear on whether the statute will ultimately survive.
The Texas Court of Criminal Appeals' decision in Ex parte Jones, 596 S.W.3d 257 (Tex. Crim. App. 2020), struck the prior version of § 21.16 as facially overbroad under the First Amendment. The court held that the statute reached a substantial amount of constitutionally protected speech — including, for example, the disclosure of intimate imagery in contexts not involving harm to the depicted individual, where the discloser lacked specific intent to harm, or where the imagery was disclosed in a journalistic or artistic context. The court emphasized that the statute's lack of a specific-intent requirement and its broad definition of the prohibited conduct created an overbreadth problem that could not be cured by case-by-case application. The legislature responded by redrafting § 21.16 to add a specific-intent-to-harm element and to narrow the conduct prong; the current § 21.16 has been the subject of subsequent challenges but has not been struck on overbreadth grounds.
Section 21.165 imported the post-Jones narrowing structure into the synthetic-media context. The statute includes the specific-intent requirement and the consent element from § 21.16, and it adds the technical specificity of the "deep fake" technology requirement. Whether these features are sufficient to survive overbreadth analysis as applied to synthetic media has not yet been decided by Texas appellate courts. The synthetic-media context raises distinct First Amendment concerns: artistic, satirical, parodic, and political uses of synthetic-imagery technology have substantial First Amendment value, and a statute that sweeps too broadly into those uses faces the same overbreadth problem that struck the prior § 21.16. The defense argument in any § 21.165 prosecution should preserve the overbreadth challenge for appellate review; this is the most likely path to a structural defense win on direct appeal.
The U.S. Supreme Court's decision in United States v. Hansen, 599 U.S. 762 (2023), refined the overbreadth doctrine in a way that bears directly on § 21.165 challenges. Hansen held that a statute is facially overbroad only if it punishes a "substantial" amount of protected speech "relative to the statute's plainly legitimate sweep." This relative-substantiality standard is more demanding than some prior formulations and may make overbreadth challenges harder to sustain. The defense argument is that the legitimate sweep of § 21.165 — punishing nonconsensual targeted-harm synthetic imagery — is narrow, and the protected-speech sweep — artistic, satirical, political, and journalistic synthetic imagery — is substantial in relative terms. The State counterargument is that the specific-intent element narrows the statute to the precise harm the legislature targeted and leaves protected speech untouched. The analysis will turn on how courts read the intent element in practice and whether prosecutorial discretion has expanded the statute beyond its narrowly drafted text.
Beyond Jones and Hansen, three foundational First Amendment decisions inform § 21.165 challenges. United States v. Stevens, 559 U.S. 460 (2010), held that the government may not categorically exclude entire categories of speech from First Amendment protection on the basis that the speech is "unprotected" — content-based speech restrictions require strict scrutiny. Brown v. Entertainment Merchants Assn., 564 U.S. 786 (2011), applied Stevens to video-game content and held that content-based restrictions on speech to minors require the same strict-scrutiny analysis as restrictions on speech to adults. United States v. Alvarez, 567 U.S. 709 (2012), struck the Stolen Valor Act and held that false statements alone are not categorically unprotected. All three decisions counsel against expansive interpretations of § 21.165 that would reach beyond the narrow targeted-harm conduct the legislature aimed at. Reno v. ACLU, 521 U.S. 844 (1997), the foundational decision on content-based regulation of internet speech, also bears on § 21.165's broad reach into electronic distribution.
Local DFW practice on emerging synthetic-media cases
DFW district attorney offices have charged § 21.165 cases sporadically since the September 1, 2023 effective date. Enforcement is early, the case law is thin, and the practice is shaped by victim NGO involvement, technology-attribution investigations, and the still-developing forensic-imagery field.
Enforcement of § 21.165 in the DFW district attorney offices (Collin, Dallas, Denton, Tarrant) has been sporadic since the September 1, 2023 effective date. The early cases have tended to involve clear targeted-harm fact patterns — typically estranged partners, schoolmate revenge scenarios, or workplace-grievance contexts where the intent element is straightforwardly satisfied by contemporaneous communications. The early enforcement pattern reflects prosecutorial caution in a new statutory area: prosecutors prefer the strong-intent cases to develop initial case law and build practice familiarity, leaving the harder-edged First Amendment cases for later in the doctrine's development. Defense counsel in DFW should expect that any § 21.165 case brought in the first wave of prosecutions will involve facts the State considers strong; the constitutional and factual defenses are correspondingly more important.
Victim NGO involvement is a distinctive feature of synthetic-media cases. Organizations including the Cyber Civil Rights Initiative, Without My Consent, and the National Center for Missing & Exploited Children (where the depicted individual is a minor) regularly involve themselves in nonconsensual intimate imagery cases, providing victim advocacy, technical takedown assistance, and policy support. Defense counsel should expect that any § 21.165 complainant has likely received NGO support during the early stages of the case, and that the complainant's statements and conduct have been shaped by that support. This affects both pretrial diversion negotiation (NGOs typically oppose diversion outcomes that do not include affirmative restorative-justice components) and trial strategy (NGO-supported complainants typically present as more confident and consistent than complainants navigating the criminal process alone).
Technology-attribution investigations are central to DFW practice on these cases. Where the State seeks to prove the defendant created the synthetic material, the investigation typically focuses on (1) device forensics — examining the defendant's computer, phone, or cloud storage for generative-AI software installations, model files, training data, prompt history, and output artifacts; (2) account attribution — linking the relevant online accounts to the defendant through IP logs, payment information, and account-creation metadata; (3) image-analysis forensics — examining the synthetic output for technical fingerprints that link it to specific generative-AI tools or to specific source images known to be in the defendant's possession. Defense forensic-imagery experts examine the same evidence and routinely identify gaps in the State's attribution case. The forensic-imagery field is still developing; defense experts in the DFW area are limited but available through national networks.
The state-jail-felony grade and non-registration default shape every DFW plea negotiation in this space. A defendant facing a clean § 21.165 charge without parallel § 22.011 or § 43.26 exposure has meaningful leverage to negotiate a deferred-adjudication outcome that avoids final conviction and preserves the non-registration baseline. Defendants facing parallel charges that trigger registration cannot achieve that outcome on the parallel charge but may still resolve the § 21.165 component in a manner that does not add registration exposure. The DFW DA offices have shown some willingness to negotiate diversion outcomes in cases where the defendant has no prior history, takes responsibility, and engages in restorative-justice steps including platform takedowns, content destruction, and victim restitution — but these outcomes are not universal and depend on the assigned prosecutor, the assigned judge, and the strength of the victim-advocacy position.
When to retain counsel
Retain experienced criminal-defense counsel immediately upon any investigation related to deepfake intimate imagery. The technical complexity, the parallel federal exposure, the platform-takedown coordination, and the emerging case law all require specialist counsel work from the earliest possible moment.
Anyone aware that they are under investigation for conduct potentially violating § 21.165 should retain experienced criminal-defense counsel immediately — before any law-enforcement interview, before any platform takedown is initiated, and before any device or account is voluntarily surrendered. The Fifth Amendment protects against compelled self-incrimination, but it does not protect against voluntary statements made before counsel is retained. Suspects who speak to investigators without counsel routinely make statements that become the State's strongest evidence on the specific-intent element — explaining why they created or distributed the material, identifying the depicted individual, characterizing the conduct as "just a joke" or "she had it coming" in ways that the State will use against them at trial. Early counsel involvement prevents these preventable disclosures.
The technical complexity of synthetic-media cases makes specialist counsel particularly important. A defense attorney unfamiliar with generative-AI tooling, with the forensic-imagery field, with the recent state and federal legislative landscape, and with the case law on First Amendment overbreadth and intent-to-harass statutes is at a structural disadvantage in this practice area. Defendants should ask their prospective counsel directly about case-specific synthetic-media experience and about familiarity with the related federal statutes (15 U.S.C. § 6851, 18 U.S.C. § 2261A, TAKE IT DOWN Act, prospective DEFIANCE Act). Counsel without this background can develop it for a specific case but the learning curve is meaningful and the time pressure in active investigations rarely permits it.
Platform-takedown coordination is a parallel concern. The TAKE IT DOWN Act creates federal platform obligations to remove nonconsensual intimate imagery within 48 hours of victim notification. Where the defendant or someone associated with the defendant maintains custody of synthetic material, prompt voluntary takedown can sometimes affect the criminal-case posture — both by demonstrating absence of ongoing intent to harm and by removing material that would otherwise be admissible evidence of the offense. Counsel must coordinate any takedown carefully: indiscriminate destruction of digital evidence may constitute spoliation or obstruction, and defense counsel must structure the takedown to comply with discovery preservation obligations while still demonstrating remedial conduct. This is a fact-specific calibration that requires experienced judgment.
The collateral consequences of any § 21.165 conviction extend well beyond the state-jail-felony sentence and fine. Background checks for employment, housing, professional licensing, and educational programs will surface any § 21.165 conviction or deferred adjudication. While the non-registration default under art. 62.001 means the defendant typically avoids the sex-offender registry, the conviction record itself carries social and professional stigma. Defense counsel must therefore think beyond the sentencing exposure to the broader collateral consequences and structure the resolution — wherever possible — to preserve professional licensing, immigration status, and employment options. Deferred adjudication, motion for non-disclosure under Texas Gov't Code ch. 411, and (where eligible) expunction are the primary tools for collateral-consequence mitigation. Engagement with these tools begins at intake, not at sentencing.
