Ustahimilivu katika enzi ya AI

Shirika la OpenAI lipo ili kuhakikisha kwamba akili unde ya jumla inanufaisha binadamu wote.

Tumekuwa tukifanya kazi bila kuchoka ili kuendana na maendeleo ya haraka ya AI yenyewe. Mnamo Aprili, Taasisi ilitangaza⁠ ruzuku zetu za kwanza zenye jumla ya dola milioni 100 za Marekani katika Sayansi ya Maisha na Kutibu Magonjwa, ikiwa na lengo la kusaidia kuzuia na kutibu magonjwa kama vile ugonjwa wa Alzheimer kwa kutumia AI ya hali ya juu. Wiki iliyopita, tulitangaza⁠ mpango wetu wa Ajira na Mustakabali wa Kiuchumi, tukiwa na matumaini ya kuelewa na kusaidia kuunda maana ya kazi na ustawi wa kiuchumi kwa vizazi vijavyo.

Leo, tunapanua maono yetu kuhusu mpango mkubwa ujao—tukihakikisha kwamba, kadiri uwezo wa AI unavyokua, uwezo wa jamii kuitumia ipasavyo utakua kwa kasi hiyo hiyo. Tunaiita kazi hii ustahimilivu wa AI: mbinu ya mfumo ikolojia inayohitajika kupunguza hatari za AI, ili jamii ifanikishe manufaa yake kikamilifu.

Kazi yetu tayari imeanza. Katika miezi michache tu tangu tuanze kazi yetu, Shirika linafanya kazi ili kukamilisha zaidi ya dola milioni 130 katika ruzuku kwa mashirika kupitia mpango wetu wa Ustahimilivu wa AI, ambazo zitatangazwa hadharani hivi karibuni na nyingine zaidi zitafuata.^1⁠

Muundo wa teknolojia zinazoleta mageuzi

Umuhimu wa ustahimilivu wa AI unaeleweka vyema zaidi kupitia mtazamo wa teknolojia za zamani ambazo ziliunda kwa kiasi kikubwa historia ya binadamu.

Kila baada ya muda fulani, teknolojia inayoibadilisha jamii kuanzia misingi yake huibuka. Wanauchumi huziita hizi “teknolojia za matumizi ya jumla.” Moto. Mashine ya uchapishaji. Umeme. Intaneti. Kila moja ilifuata mwelekeo uliofanana: ubunifu wa kasi, hatari halisi, na taasisi zikiharakisha kuendana na kasi hiyo. Lakini kila mfano pia unaonyesha kinachohitajika ili kufanya teknolojia yenye uwezo mkubwa iwe salama.

Moto ulifanya ustaarabu wa binadamu uwezekane. Ulitupatia joto, ukapika chakula chetu, na ukatulinda dhidi ya wanyama wanaowinda. Pia iliteketeza miji yetu kabisa. Kadiri muda ulivyopita, jamii zilijenga ustahimilivu: vifaa vinavyostahimili moto, mitandao ya hidranti za zimamoto, idara za zimamoto za kitaalamu, na kanuni za ujenzi. Mfumo wa ikolojia, tabaka kwa tabaka.

Umeme ulifuata njia ile ile. Baada ya Kituo cha Pearl Street cha Edison kuangaza Manhattan mwaka 1882, umeme ulisababisha mioto, kugongwa na umeme, na taharuki kwa umma. Bila hatua za ulinzi kama vile nyaya zilizowekewa vihami, vivunja mzunguko wa umeme, na kanuni za usalama, wafanyakazi na watu waliokuwepo walikufa baada ya kugongwa na umeme katika miji kote nchini. Miji ilijadili iwapo jaribio hilo linafaa kuachwa kabisa. Badala yake, kadiri teknolojia ilivyoendelea, tulianzisha mashirika huru ya upimaji kama Underwriters Laboratories, viwango vya sekta kama National Electrical Code, na uwekezaji wa umma uliofikisha umeme kwa jamii ambazo soko lilikuwa limeziacha nyuma. Kila safu ilifanya umeme kuwa salama zaidi na kupatikana kwa urahisi zaidi; leo, umeme ni salama kiasi kwamba mtoto anaweza kuwasha swichi na mwanga ukatokea.

Hivi ndivyo ustahimilivu unavyoonekana unapotekeleza vizuri.

AI inahitaji mfumo wa ikolojia wa ustahimilivu

AI inafuata mwelekeo uleule kama teknolojia za awali, lakini inasonga kwa kasi kubwa.

Bado tuko katika hatua zake za mwanzo, lakini manufaa yake tayari yako wazi: AI inapunguza vikwazo vya kuanzisha biashara, kupanua upatikanaji wa elimu, kuharakisha ugunduzi wa kisayansi, na kubadilisha sekta ya tiba.

Wakati huohuo, hatari zinaibuka kwa kasi vivyo hivyo—na zikiwa kama taswira sawia na ya manufaa ya AI. Ukuaji huo huo unaounda sekta mpya unaweza kubadili kabisa sekta zilizopo na kuvuruga taaluma. Mifumo hiyo hiyo inayoweza kuwasaidia vijana kujifunza na kubuni inaweza pia kusababisha tabia zisizofaa. Zana zinazoharakisha utafiti wa kibiolojia zinaweza kupunguza vikwazo vya kuunda vijidudu hatari vinavyosababisha magonjwa. Na uwezo wa AI wa kuandika msimbo, ikiwa mikononi mwa watu wasiofaa, unaweza kutishia miundombinu muhimu.

Timu ya awali ya OpenAI iliamini kuwa kuhakikisha kwamba AI inanufaisha jamii kulitegemea hasa kutatua tatizo la kiufundi la upatanishaji. Hilo bado ni muhimu sana—na ni kiini cha kazi yetu—lakini sasa tunaamini kwamba ni sehemu moja tu ya fumbo. Kadiri AI inavyoenea katika sekta na mataifa, jamii pia itahitaji utafiti huru, miundombinu ya umma, uratibu wa tasnia, na nyanja mpya kabisa za utaalamu. Kwa kifupi, itahitaji ustahimilivu wa AI.

Tumeamua kuelekeza kazi yetu ya awali katika maeneo manne^2⁠ yaliyo katika makutano ya hatari kubwa zinazoweza kutokea hivi karibuni na athari za haraka:

1. Ustahimilivu wa kibaiolojia ili kusaidia kuzuia magonjwa tandavu yaliyoundwa kwa uhandisi katika siku zijazo;

2. Ustahimilivu wa kimtandao kufanya kazi ili kuhakikisha usalama wa mifumo muhimu ya dunia yetu;

3. Usalama wa miundo ya AI ili kuimarisha udhibiti wa binadamu juu ya miundo tunayounda; na

4. Athari ya AI kwa vijana ili kusaidia kuifanya teknolojia kuwa nguvu chanya kwa vizazi vijavyo.

Kazi yetu bado inaanza. Tunapanga kutoa maelezo zaidi kuhusu mikakati yetu na ruzuku za awali katika kila eneo, na kupanuka hadi maeneo mengine kadri muda unavyopita.

Bio-resilience

AI will enable biological research to move at unprecedented speed, helping develop new cures and public health improvements that enable us to all live healthier and longer. However, these same capabilities could also be misused by malicious actors, lowering the barrier to designing harmful pathogens.

The age of AI requires a renewed focus on biosecurity. Because advanced AI systems could be misused by bad actors to help create a wide range of biological threats, we will prioritize pathogen-agnostic biosecurity solutions. This will require investments across prevention, detection, and defense. We need to make it harder for malicious actors to access the expertise, equipment, and materials to create biological threats, improve our ability to identify and track novel outbreaks early, and strengthen the technologies—such as protective equipment, indoor air cleaning systems, and medical countermeasures—needed to respond quickly and effectively.

Cyber-resilience

AI has begun to rapidly reshape the cybersecurity landscape. The work that once required specialized teams can now be assisted or automated by capable models. At the same time, rapidly-improving AI capabilities can also be used to accelerate cyberdefenders, including by identifying and patching vulnerabilities and accelerating response.

Many large companies and private actors can spend heavily on cyber to secure their own systems, including with new advances in AI. We anticipate focusing significant resources on securing other important societal actors that are less resourced and will have a much harder time deploying AI-ready cyberdefenses as quickly as needed. In parallel, we are also focused on preparing for novel security challenges that artificial general intelligence will ultimately bring.

AI model safety

AI model safety focuses on the behavior of the systems themselves—whether they are truthful, reliable, and aligned with human intent. In a world where this goes awry, models can break out and behave in unpredictable ways, deceiving us or pursuing goals beyond their design. Getting this right becomes increasingly important as AI systems grow more autonomous and approach—and eventually surpass—human-level intelligence.

AI companies are investing substantial resources in model safety. However, the importance of this challenge calls for a broader, more robust ecosystem: independent institutions to evaluate model safety, public infrastructure to verify models’ safe deployment in practice, and continued advances in alignment science that advance the field broadly.

AI’s impact on young people

Young people are often the earliest adopters of new technologies, using them to learn, create, communicate, and explore the world. AI is no exception. But as these tools become an increasing part of young people’s daily lives, it is critical that we develop a stronger evidence base to understand its impacts.

Families, schools, policymakers, and community organizations are all grappling with questions about how and when young people engage with AI—including its impact on human connection, learning, and development. Our initial focus will be on advancing independent research to help guide those decisions—to better understand where AI can support development, the risks it may introduce, and the contexts that shape those effects.

These insights should drive broad safety standards and design principles that guide how any AI product is developed, how schools choose to deploy them, and if and how families decide to incorporate these technologies into their lives.

The work ahead

There is one critical difference between AI and the technologies that came before it: speed.

Fire resilience took millennia. Electricity resilience took decades. AI resilience is evolving in a matter of years. The systems that make it safe, reliable, and broadly beneficial must be built alongside it.

If we get it right, AI can become part of the foundational infrastructure of modern life—expanding access to knowledge, accelerating discovery, and improving lives at a global scale.

But that outcome isn’t guaranteed. No general purpose technology ever made itself safe.

Resilience is a permanent discipline that requires many people and institutions to build, invest, and collaborate. That is the work ahead, and it is one of the defining challenges of our time. We hope you’ll join us.