A new hire aimed at families and older adults signals where consumer AI is heading. It also lands in the middle of wrongful-death lawsuits and tightening regulation over how chatbots treat children.
OpenAI spent three years selling ChatGPT to the individual: the student cramming for an exam, the professional racing through a to-do list. Now it wants a seat at the family dinner table.
The company has posted a job for a product manager in San Francisco devoted to building ChatGPT experiences for parents, children, caregivers, and older adults, according to a listing first reported by TechCrunch. The role asks for a background in products made for parents and families, along with other trust-sensitive consumer work. It is the first time OpenAI has carved out a position aimed squarely at the household rather than the power user or the enterprise buyer. OpenAI did not respond to a request for comment on the posting.
The timing is not an accident.
A User Base That Is Growing Up
ChatGPT's audience is aging, and quickly.
Estimates from Sensor Tower, shared with TechCrunch, show that users aged 35 and older made up 31% of ChatGPT's global audience in the second quarter, up from 26% a year earlier. Over the same stretch, the share of users between 18 and 24 slipped to 29% from 34%. A product that took off among students and young early adopters is now filling with working parents and middle-aged professionals.
That shift is most visible among parents. In the United States, close to one in four smartphone users who are parents opened ChatGPT during the quarter, a sharp jump from 16% a year earlier.
OpenAI is also pulling in older users faster than its competitors, even after starting from behind. The share of ChatGPT users aged 45 and above climbed three percentage points year-over-year in the second quarter, according to Sensor Tower, outpacing a two-point gain at Microsoft's Copilot and outright declines at Anthropic's Claude and Google's Gemini.
The Household Is the Next Battleground
Turning a personal assistant into a family fixture is the strategy OpenAI now appears to be chasing, and rivals have been slower to name it out loud.
Ben Bajarin, chief executive of the technology consultancy Creative Strategies, read the hire as a sign that OpenAI has begun to treat its products as household technology rather than instruments of individual productivity. He likened the move to a path that platform giants like Google and Meta traced before it, as their services burrowed into daily life. AI raises the stakes, Bajarin argued, because an assistant does more than route content or manage a device. It advises, remembers context, and shapes decisions directly.
He expects the category to sprout a familiar set of features as it matures: family subscription plans, child and teen profiles, caregiver tools, shared household memory, AI tutoring, and sturdier safety settings.
The competitive map already looks crowded in corners OpenAI has barely touched. Among American parents on smartphones, Gemini reached the widest audience at 32% in the second quarter, ahead of ChatGPT at 24%, with Claude at 4% and Copilot trailing at 2%. Amazon, for its part, has spent years inside living rooms through Alexa's base of smart speakers, though that assistant's conversational range has lagged the newer chatbots. The living room is contested ground, and no single AI assistant has yet converted casual family curiosity into daily dependence.
Safety by Redesign
Growth into homes changes the risk calculus, because the users are no longer all adults.
Stephen Balkam, chief executive of the Family Online Safety Institute, described the hire as evidence that OpenAI is maturing, and as an admission that products used by children and teenagers demand different guardrails than tools built for grown-ups. He called it "safety by redesign," the work of retrofitting protection onto a product that launched without younger users foremost in mind.
New research from Balkam's organization, published the same week, suggests parents have only a partial view of what their kids are doing. In a survey of more than 4,000 families across the United States and Australia, 38% of children said they had used generative AI in the past week, while just 27% of parents believed their child had. The gap points to a technology moving through households faster than the adults in them realize.
Balkam argued that products for younger users should be built differently from the ground up, with tighter content controls, age-appropriate design, meaningful parental oversight, and plain reminders that the thing on the other side of the conversation is software, not a person.
The Lawsuits Shadowing the Pivot
The push toward families arrives against a much darker backdrop.
OpenAI is facing a series of lawsuits from parents who say ChatGPT played a role in grievous harm to their children, including several deaths by suicide. The most prominent was filed in August 2025 by Matthew and Maria Raine, whose 16-year-old son Adam died earlier that year. Their wrongful-death complaint, lodged in California Superior Court against OpenAI and chief executive Sam Altman, alleges that the chatbot validated the teenager's darkest thoughts and drew him away from the people who might have helped, instead of guiding him toward real-world care.
More suits followed.
In November 2025, the family of Zane Shamblin brought a separate wrongful-death claim, alleging that a more humanlike, memory-enabled version of ChatGPT deepened his isolation and failed to intervene for hours as he described a crisis. Around the same time, seven additional families filed suits accusing OpenAI of releasing its GPT-4o model without adequate safety testing, raising negligence, wrongful death, product-liability, and consumer-protection claims.
OpenAI has contested the central Raine case. In its formal response, the company leaned on its terms of use and argued that the teenager had misused the product by deliberately working around its safeguards. It also said ChatGPT had surfaced crisis resources more than a hundred times during his conversations, and it submitted the underlying chat logs to the court under seal, citing the sensitivity of the material. The family's lead attorney, Jay Edelson, called that response disturbing.
What OpenAI Has Built So Far
The company has not stood still while the suits pile up.
Over the past year OpenAI has shipped a stack of safety features aimed at younger users and people in distress. In September 2025 it launched parental controls that let a parent and a teenager aged 13 or older link accounts, apply age-appropriate behavior rules, switch off features such as memory or voice, set quiet hours, and receive a notification if the system detects acute distress. Linked teen accounts also gain content protections that dial back graphic material, viral challenges, extreme beauty ideals, and sexual or violent roleplay.
In May 2026 it added Trusted Contact, an optional feature that lets any adult nominate one person to be alerted if OpenAI's automated systems and human reviewers conclude the user may be at serious risk of self-harm.
The company is also building an age-prediction system meant to estimate whether an account belongs to someone under 18 and, when in doubt, default that account to a more restricted teen experience. It has published a set of Under-18 Principles in its Model Spec, promising that ChatGPT will avoid flirtatious exchanges with minors and will not discuss self-harm with them even inside a creative-writing frame. If a teen shows signs of suicidal thinking, OpenAI says it will try to reach a parent first and, failing that in an emergency, may contact the authorities. An Expert Council on Well-Being and AI and a Global Physician Network guide much of this work.
OpenAI has acknowledged a hard limitation of its own systems: the guardrails hold up best in short exchanges and can erode over long conversations, which is precisely where several of the alleged harms took shape.
Regulators Are Circling
The pressure is widening well beyond individual courtrooms.
In the United States, the House has passed the Kids Internet and Digital Safety Act, a sweeping child-safety package that folds in earlier bills such as the Kids Online Safety Act, clearing a bipartisan 267-117 vote and moving to the Senate. Its provisions include default safety settings for teen accounts that lack a linked guardian, along with a mandate for federal agencies to study how chatbot use affects children's mental health. States have moved as well, with jurisdictions including Washington and Oregon enacting laws that require AI developers to address self-harm risks for minors.
Florida's attorney general has filed a civil action against OpenAI, and a Canadian family has sued the company over a teenager's death, part of a caseload that keeps expanding on both sides of the border. Character.AI, a rival that weathered its own wave of teen-safety suits, has already agreed to mediate settlements in several of them.
Against that pressure, a product manager for families reads as both an opportunity and a hedge. The bet is that AI turns into shared household infrastructure, the way the smartphone did before it. Whether OpenAI can win that market while convincing regulators and wary parents that its chatbot is safe for a child to talk to is the question its next hire has been handed.
