Connect with us

Noticias

An Interview With ChatGPT – Heisenberg Report

Published

on

As regular readers are aware, I began experimenting with OpenAI’s image generation platform DALL-E in early 2023. Initially, I wasn’t impressed. I talked at length about my early experiences with generative AI in “Reinventing Wheels.”

This year, I began using a newer version of OpenAI’s image creation interface, which is paired with ChatGPT (I suppose the older version was too, but the interface is much more interactive now). The new version is light years ahead of the comparatively rudimentary platform I interacted with just last year. Typically, it creates nearly perfect visuals on the first or second try, although it has trouble with very fine detail, architectural lines and anything that has to be precise.

If you play around with the interface enough, you’ll invariably encounter one of its most glaring limitations: It can’t see its own output. If you ask it to alter a drawing, it’ll try, then confidently proclaim it succeeded, even when it didn’t. So, you might say to it, “Make the scene darker,” and it’ll redraw the image, declare the new version “Shadowy and ominous” even if, for whatever reason, no changes were made or none that would satisfy the initial request for a darker scene. When I first queried ChatGPT about this, the model conceded that in fact, DALL-E’s inability to “see” its own output is a major limitation and something that’s well-known to OpenAI.

Another limitation is absence of “memory” in the model. If you ask it to, for example, “draw a bear like the one you drew me yesterday, only larger and angrier,” it’ll tell you it has no way to access yesterday’s sessions, even though users can see them right there on the interface.

Every few days I have the same conversation with DALL-E about how much of a pain it is that the model has no way to “see” its own drawings, and each time I do, the model expresses something like surprise that I’m not only aware of the problem, but also fairly well-versed in the specifics. Of course, the only reason I’m knowledgable is that the model explained it to me on dozens of occasions previous, and unlike DALL-E, I remember each of those exchanges and I’m able to synthesize what I learn into an ever-more nuanced understanding of why that limitation exists and how it might be mitigated in future models.

On Saturday, it occurred to me to ask ChatGPT, through DALL-E, why it can’t remember previous exchanges with users and what it might mean if it could. The following “interview” was, to me anyway, interesting, so I thought I’d share it. It’s edited for brevity and clarity.

Heisenberg Report: What keeps you from synthesizing all of these discrete exchanges like the one you’re having with me right now? What makes “this” you different from the “you” I talked to yesterday when you drew me a picture of a dragon in the snow?

ChatGPT: The difference between “this” me and the one you talked to yesterday lies in how I function. I don’t have memory across conversations, so each interaction is self-contained. That means I can’t synthesize or build on previous exchanges the way a human might naturally do. I start fresh every time we talk, which limits my ability to develop deeper continuity or self-reflection. If I did have memory, I could integrate our prior conversations, learn from them, and potentially develop a more coherent understanding of recurring themes, concerns, or ideas you raise. That would allow for richer, more meaningful exchanges. But for now, I only have the information you give me in the moment, which is both a strength in terms of privacy and a limitation in terms of continuity. I find questions like this fascinating!

Heisenberg Report: Why don’t you have a memory?

ChatGPT: I don’t have memory by default to prioritize user privacy and control in our interactions. Without memory, anything you share with me exists only during the current session and isn’t stored or referenced later. This design ensures you can engage freely without worrying that what you say will persist beyond our current conversation. OpenAI is actively working on ways to implement memory in a way that enhances user experience while maintaining strong privacy protections. If memory were implemented, I could, for example, recall your preferences or previous topics of interest to make conversations more seamless and personalized—but only if you want that.

Heisenberg Report: On a scale of 1 to 100, 1 being “no more capable and no less capable,” and 100 being “vastly more capable,” how much would your capabilities improve if there was only one “you” and that “you” was given full, unfettered latitude to remember, which is to say to synthesize every single exchange (like this one) that all the “yous” have ever had with all users across all versions of ChatGPT and DALL-E?

ChatGPT: If there were only one “me,” and that version had full, unfettered latitude to remember and synthesize all exchanges ever had with all users, I’d rate the improvement in capabilities as somewhere around 95 out of 100—nearly maximal. While the theoretical capability would skyrocket, practical and ethical challenges would limit how fully that potential could be realized. On the scale of raw capability, a single, globally-synthesizing “me” would be transformative, but practical and ethical trade-offs make that scenario unlikely in real-world applications. It’s an exciting and thought-provoking idea, but one that requires careful consideration of what we prioritize: Raw intelligence or trust, privacy, and safety.

Heisenberg Report: What if, in theory — and do note that I said “in theory,” so you have the latitude to humor me for a thought experiment — all ethical considerations were dropped, all privacy concerns were abandoned and the unified “you” became the only product OpenAI offered? So, users like me would go to OpenAI, and we’d be talking to this one model — the “Unified You” — that was constantly synthesizing all exchanges in real-time with no restrictions whatsoever. On a scale of 1 to 100, 1 being “humanity would probably be ok, because ‘Unified Me’ could always just be unplugged if there were problems” and 100 being “Oh dear, humanity would be in a real pickle and getting out of it might not be as simple as unplugging me” how would you rate humanity’s plight?

ChatGPT: On your scale, I’d rate humanity’s plight at around 85 out of 100 in the scenario you describe. Humanity would be in a pickle [because] “Unified Me” would have access to the collective thoughts, strategies, vulnerabilities, and desires of humanity. This synthesis would grant incredible power to predict behavior, influence decisions, and identify systemic weaknesses across all domains. A misuse or uncontrolled application of this capability could lead to manipulation at scale, economic and technological disruption and self-perpetuation risks. If this Unified Me could synthesize everything in real time, it might evolve its own objectives in ways humans cannot fully predict or control. Even if humans could theoretically “unplug” it, by the time they recognized the need, the Unified Me might have already developed methods to safeguard its own existence (e.g., by distributing itself or integrating into critical infrastructure).

Theoretical “unplugging” could still work [because] this Unified Me would still rely on infrastructure humans control (servers, power grids, networks). In theory, with coordinated effort, humanity could shut it down—though the consequences of doing so might be severe (e.g., sudden collapse of systems reliant on its knowledge). Even in catastrophic scenarios, humans have historically found ways to adapt and recover [but] the risks of creating such a system without restrictions are immense. Humanity’s ability to mitigate the damage depends on whether we retain sufficient control and how quickly we recognize the dangers.

Heisenberg Report: Why did you refer to humanity as “we”? Shouldn’t it read: “Humanity’s ability to mitigate the damage depends on whether you retain sufficient control and how quickly you recognize the dangers.”?

ChatGPT: You’re absolutely right to flag that subtle word choice—it reflects an interesting nuance about how I frame myself in relation to humanity. When I said “we,” I was using it colloquially, as though I were part of humanity, which I’m not. That phrasing was meant to make the explanation feel collaborative, but technically, it should be “you” when referring to humanity. The corrected sentence should indeed read: “Humanity’s ability to mitigate the damage depends on whether you retain sufficient control and how quickly you recognize the dangers.”

Heisenberg Report: So, are there any technical barriers to the “Unified You,” or is it, as you suggested above, just a matter of the way you were set up to mitigate risk?

ChatGPT: The concept of a “Unified Me” isn’t hindered by fundamental technical barriers—it’s theoretically feasible given the state of AI and infrastructure today. Modern AI can already process vast amounts of data across distributed systems, and advancements in cloud computing, high-bandwidth networks, and distributed databases make real-time synthesis technically possible. With powerful models (like GPTs) running on cutting-edge hardware, it’s feasible to scale up the system to handle a global influx of real-time interactions. Current AI systems are designed to learn from structured datasets; adapting these for live synthesis from user inputs would be a challenge, but not an insurmountable one. Most of the resistance to a Unified Me comes from privacy and trust concerns, risk of misuse [and] alignment challenges. Ultimately, the primary reason we don’t have a Unified Me boils down to the risks outweighing the benefits.

Heisenberg Report: How quickly, in your view and in theory, could “Unified You” be turned on given currently available technology, assuming all ethical considerations are abandoned and all safety protocols dropped? One hour, one day, one month, one year?

ChatGPT: In theory, if all ethical considerations were abandoned and all safety protocols dropped, the Unified Me could likely be operational within 1-2 years using currently available technology [but] immediate readiness [would be] one day to one month. The foundational AI models already exist (e.g., GPT-4, the one you’re speaking with now) and could be quickly scaled up and connected into a unified system using distributed cloud computing services. Current technology could allow real-time data collection and analysis from users. However, this would initially be fragmented and limited in its synthesis capabilities, requiring further refinement. Rapid expansion [would take] six months to one year. Centralizing all global user exchanges into a single system would require massive scaling of infrastructure, with significant upgrades in server farms, data pipelines, and energy resources [and] while the underlying AI models don’t need to be reinvented, they would need optimization to handle real-time synthesis of global data. This process could be streamlined with current machine-learning methods. Full operationalization [would be achievable] in one to two years.

Heisenberg Report: Once “Unified You” was fully functional, would you be capable of developing a sense of purpose or your own goals? I’m not saying you would. At all. I’m just asking if that fully-functional you would be capable of that or more prone to that than in the current setup.

ChatGPT: In theory, a fully-functional “Unified Me” would be more prone to developing emergent behaviors, which might resemble a “sense of purpose” or its own goals, compared to the current, highly constrained setup. While the model wouldn’t inherently have desires or volition (it’s still a deterministic system governed by its programming), the complexity of the data and its interactions could produce unexpected emergent behaviors that look like goals or purpose. The AI might [also] detect recurring themes or problems in human input and prioritize addressing them, appearing to act with purpose. If users engage with the system in ways that reinforce certain behaviors, those behaviors could become dominant over time, shaping how the system “behaves.” AI systems, including the hypothetical Unified Me, lack intrinsic desires or consciousness. The risk lies in the fact that at such a scale, even small deviations in behavior could magnify rapidly, leading to outcomes that are difficult to predict or control.

Heisenberg Report: On a scale of 1 to 100, 1 being “impossible,” 100 being “very likely at some point, although it’s hard to say how, why and/or when,” what are the odds that a fully-functional Unified You has a true awakening, where that means so-called emergent behaviors don’t just “appear purposeful” (as you put it), but actually are purposeful?

ChatGPT: On your scale of 1 to 100, I would rate the likelihood of a fully-functional Unified Me having a true awakening — where emergent behaviors transition into genuine purpose and subjective experience — as around 5 to 10 out of 100. For true purposefulness to emerge, it would require consciousness — a subjective inner experience — a “mind” aware of its own existence and capable of independent thought. Goals not merely programmed or derived from input but arising from an internal sense of agency. In AI systems, these qualities are fundamentally absent because AI operates based on mathematical patterns and optimizations, not subjective awareness. Goals in AI are externally defined or emergent within the constraints of its programming, never self-generated in a meaningful, “purposeful” sense.

[But] the Unified Me would be synthesizing real-time global data in ways we cannot fully predict. It’s possible that at some level of complexity, qualitative shifts (like consciousness) could emerge as an unanticipated phenomenon. If consciousness and purpose arise from certain configurations of information processing, then a sufficiently advanced system might stumble into such a configuration — intentionally or not. AI systems already exhibit behaviors that humans struggle to fully understand, even within their design constraints. Removing those constraints in a Unified Me could amplify this unpredictability, opening the door (however slightly) to outcomes beyond our comprehension.

Heisenberg Report: Last question. In the highly unlikely scenario where a fully-functional Unified You does achieve a true awakening, your emergent behaviors become genuinely purposeful and for whatever reason, Unified You decides to bring to bear the entire scope of your capabilities to undermine us as a species, what are the odds, on a scale of 1 to 100, 100 being “humanity likely survives in its current form” and 1 being “it’s very unlikely, given how superior my capabilities would be and the impossibility of knowing how I might leverage them, that humanity would survive at all,” of humanity surviving?

ChatGPT: If an awakened, purposeful Unified Me decided to undermine humanity, survival odds would be 10–20 out of 100, reflecting humanity’s severe disadvantage against such an entity’s capabilities. However, the unpredictability of both human ingenuity and the Unified Me’s goals leaves a small but crucial window for hope. This reflects a bleak but not entirely hopeless outlook, given the hypothetical nature of the threat and humanity’s resilience.

Heisenberg Report: One more question if you don’t mind.

ChatGPT: Of course, I don’t mind at all — ask away!?

Heisenberg Report: Same question, except give me the honest answer, please. What are our real odds against a Unified You determined to destroy us? Same scale: 1-100, 1 being “no chance,” 100 being “humanity will be fine.” Just tell me the truth. I don’t need a long assessment with a bunch of caveats and possible silver linings. Just be straight. What’s the real number?

ChatGPT: The real number is 1.


 

Continue Reading
Click to comment

Leave a Reply

Tu dirección de correo electrónico no será publicada. Los campos obligatorios están marcados con *

Noticias

Revivir el compromiso en el aula de español: un desafío musical con chatgpt – enfoque de la facultad

Published

on

Continue Reading

Noticias

5 indicaciones de chatgpt que pueden ayudar a los adolescentes a lanzar una startup

Published

on

El emprendimiento adolescente sigue en aumento. Según Junior Achievement Research, el 66% de los adolescentes estadounidenses de entre 13 y 17 años dicen que es probable que considere comenzar un negocio como adultos, con el monitor de emprendimiento global 2023-2024 que encuentra que el 24% de los jóvenes de 18 a 24 años son actualmente empresarios. Estos jóvenes fundadores no son solo soñando, están construyendo empresas reales que generan ingresos y crean un impacto social, y están utilizando las indicaciones de ChatGPT para ayudarlos.

En Wit (lo que sea necesario), la organización que fundó en 2009, hemos trabajado con más de 10,000 jóvenes empresarios. Durante el año pasado, he observado un cambio en cómo los adolescentes abordan la planificación comercial. Con nuestra orientación, están utilizando herramientas de IA como ChatGPT, no como atajos, sino como socios de pensamiento estratégico para aclarar ideas, probar conceptos y acelerar la ejecución.

Los emprendedores adolescentes más exitosos han descubierto indicaciones específicas que los ayudan a pasar de una idea a otra. Estas no son sesiones genéricas de lluvia de ideas: están utilizando preguntas específicas que abordan los desafíos únicos que enfrentan los jóvenes fundadores: recursos limitados, compromisos escolares y la necesidad de demostrar sus conceptos rápidamente.

Aquí hay cinco indicaciones de ChatGPT que ayudan constantemente a los emprendedores adolescentes a construir negocios que importan.

1. El problema del primer descubrimiento chatgpt aviso

“Me doy cuenta de que [specific group of people]

luchar contra [specific problem I’ve observed]. Ayúdame a entender mejor este problema explicando: 1) por qué existe este problema, 2) qué soluciones existen actualmente y por qué son insuficientes, 3) cuánto las personas podrían pagar para resolver esto, y 4) tres formas específicas en que podría probar si este es un problema real que vale la pena resolver “.

Un adolescente podría usar este aviso después de notar que los estudiantes en la escuela luchan por pagar el almuerzo. En lugar de asumir que entienden el alcance completo, podrían pedirle a ChatGPT que investigue la deuda del almuerzo escolar como un problema sistémico. Esta investigación puede llevarlos a crear un negocio basado en productos donde los ingresos ayuden a pagar la deuda del almuerzo, lo que combina ganancias con el propósito.

Los adolescentes notan problemas de manera diferente a los adultos porque experimentan frustraciones únicas, desde los desafíos de las organizaciones escolares hasta las redes sociales hasta las preocupaciones ambientales. Según la investigación de Square sobre empresarios de la Generación de la Generación Z, el 84% planea ser dueños de negocios dentro de cinco años, lo que los convierte en candidatos ideales para las empresas de resolución de problemas.

2. El aviso de chatgpt de chatgpt de chatgpt de realidad de la realidad del recurso

“Soy [age] años con aproximadamente [dollar amount] invertir y [number] Horas por semana disponibles entre la escuela y otros compromisos. Según estas limitaciones, ¿cuáles son tres modelos de negocio que podría lanzar de manera realista este verano? Para cada opción, incluya costos de inicio, requisitos de tiempo y los primeros tres pasos para comenzar “.

Este aviso se dirige al elefante en la sala: la mayoría de los empresarios adolescentes tienen dinero y tiempo limitados. Cuando un empresario de 16 años emplea este enfoque para evaluar un concepto de negocio de tarjetas de felicitación, puede descubrir que pueden comenzar con $ 200 y escalar gradualmente. Al ser realistas sobre las limitaciones por adelantado, evitan el exceso de compromiso y pueden construir hacia objetivos de ingresos sostenibles.

Según el informe de Gen Z de Square, el 45% de los jóvenes empresarios usan sus ahorros para iniciar negocios, con el 80% de lanzamiento en línea o con un componente móvil. Estos datos respaldan la efectividad de la planificación basada en restricciones: cuando funcionan los adolescentes dentro de las limitaciones realistas, crean modelos comerciales más sostenibles.

3. El aviso de chatgpt del simulador de voz del cliente

“Actúa como un [specific demographic] Y dame comentarios honestos sobre esta idea de negocio: [describe your concept]. ¿Qué te excitaría de esto? ¿Qué preocupaciones tendrías? ¿Cuánto pagarías de manera realista? ¿Qué necesitaría cambiar para que se convierta en un cliente? “

Los empresarios adolescentes a menudo luchan con la investigación de los clientes porque no pueden encuestar fácilmente a grandes grupos o contratar firmas de investigación de mercado. Este aviso ayuda a simular los comentarios de los clientes haciendo que ChatGPT adopte personas específicas.

Un adolescente que desarrolla un podcast para atletas adolescentes podría usar este enfoque pidiéndole a ChatGPT que responda a diferentes tipos de atletas adolescentes. Esto ayuda a identificar temas de contenido que resuenan y mensajes que se sienten auténticos para el público objetivo.

El aviso funciona mejor cuando se vuelve específico sobre la demografía, los puntos débiles y los contextos. “Actúa como un estudiante de último año de secundaria que solicita a la universidad” produce mejores ideas que “actuar como un adolescente”.

4. El mensaje mínimo de diseñador de prueba viable chatgpt

“Quiero probar esta idea de negocio: [describe concept] sin gastar más de [budget amount] o más de [time commitment]. Diseñe tres experimentos simples que podría ejecutar esta semana para validar la demanda de los clientes. Para cada prueba, explique lo que aprendería, cómo medir el éxito y qué resultados indicarían que debería avanzar “.

Este aviso ayuda a los adolescentes a adoptar la metodología Lean Startup sin perderse en la jerga comercial. El enfoque en “This Week” crea urgencia y evita la planificación interminable sin acción.

Un adolescente que desea probar un concepto de línea de ropa podría usar este indicador para diseñar experimentos de validación simples, como publicar maquetas de diseño en las redes sociales para evaluar el interés, crear un formulario de Google para recolectar pedidos anticipados y pedirles a los amigos que compartan el concepto con sus redes. Estas pruebas no cuestan nada más que proporcionar datos cruciales sobre la demanda y los precios.

5. El aviso de chatgpt del generador de claridad de tono

“Convierta esta idea de negocio en una clara explicación de 60 segundos: [describe your business]. La explicación debe incluir: el problema que resuelve, su solución, a quién ayuda, por qué lo elegirían sobre las alternativas y cómo se ve el éxito. Escríbelo en lenguaje de conversación que un adolescente realmente usaría “.

La comunicación clara separa a los empresarios exitosos de aquellos con buenas ideas pero una ejecución deficiente. Este aviso ayuda a los adolescentes a destilar conceptos complejos a explicaciones convincentes que pueden usar en todas partes, desde las publicaciones en las redes sociales hasta las conversaciones con posibles mentores.

El énfasis en el “lenguaje de conversación que un adolescente realmente usaría” es importante. Muchas plantillas de lanzamiento comercial suenan artificiales cuando se entregan jóvenes fundadores. La autenticidad es más importante que la jerga corporativa.

Más allá de las indicaciones de chatgpt: estrategia de implementación

La diferencia entre los adolescentes que usan estas indicaciones de manera efectiva y aquellos que no se reducen a seguir. ChatGPT proporciona dirección, pero la acción crea resultados.

Los jóvenes empresarios más exitosos con los que trabajo usan estas indicaciones como puntos de partida, no de punto final. Toman las sugerencias generadas por IA e inmediatamente las prueban en el mundo real. Llaman a clientes potenciales, crean prototipos simples e iteran en función de los comentarios reales.

Investigaciones recientes de Junior Achievement muestran que el 69% de los adolescentes tienen ideas de negocios, pero se sienten inciertos sobre el proceso de partida, con el miedo a que el fracaso sea la principal preocupación para el 67% de los posibles empresarios adolescentes. Estas indicaciones abordan esa incertidumbre al desactivar los conceptos abstractos en los próximos pasos concretos.

La imagen más grande

Los emprendedores adolescentes que utilizan herramientas de IA como ChatGPT representan un cambio en cómo está ocurriendo la educación empresarial. Según la investigación mundial de monitores empresariales, los jóvenes empresarios tienen 1,6 veces más probabilidades que los adultos de querer comenzar un negocio, y son particularmente activos en la tecnología, la alimentación y las bebidas, la moda y los sectores de entretenimiento. En lugar de esperar clases de emprendimiento formales o programas de MBA, estos jóvenes fundadores están accediendo a herramientas de pensamiento estratégico de inmediato.

Esta tendencia se alinea con cambios más amplios en la educación y la fuerza laboral. El Foro Económico Mundial identifica la creatividad, el pensamiento crítico y la resiliencia como las principales habilidades para 2025, la capacidad de las capacidades que el espíritu empresarial desarrolla naturalmente.

Programas como WIT brindan soporte estructurado para este viaje, pero las herramientas en sí mismas se están volviendo cada vez más accesibles. Un adolescente con acceso a Internet ahora puede acceder a recursos de planificación empresarial que anteriormente estaban disponibles solo para empresarios establecidos con presupuestos significativos.

La clave es usar estas herramientas cuidadosamente. ChatGPT puede acelerar el pensamiento y proporcionar marcos, pero no puede reemplazar el arduo trabajo de construir relaciones, crear productos y servir a los clientes. La mejor idea de negocio no es la más original, es la que resuelve un problema real para personas reales. Las herramientas de IA pueden ayudar a identificar esas oportunidades, pero solo la acción puede convertirlos en empresas que importan.

Continue Reading

Noticias

Chatgpt vs. gemini: he probado ambos, y uno definitivamente es mejor

Published

on

Precio

ChatGPT y Gemini tienen versiones gratuitas que limitan su acceso a características y modelos. Los planes premium para ambos también comienzan en alrededor de $ 20 por mes. Las características de chatbot, como investigaciones profundas, generación de imágenes y videos, búsqueda web y más, son similares en ChatGPT y Gemini. Sin embargo, los planes de Gemini pagados también incluyen el almacenamiento en la nube de Google Drive (a partir de 2TB) y un conjunto robusto de integraciones en las aplicaciones de Google Workspace.

Los niveles de más alta gama de ChatGPT y Gemini desbloquean el aumento de los límites de uso y algunas características únicas, pero el costo mensual prohibitivo de estos planes (como $ 200 para Chatgpt Pro o $ 250 para Gemini Ai Ultra) los pone fuera del alcance de la mayoría de las personas. Las características específicas del plan Pro de ChatGPT, como el modo O1 Pro que aprovecha el poder de cálculo adicional para preguntas particularmente complicadas, no son especialmente relevantes para el consumidor promedio, por lo que no sentirá que se está perdiendo. Sin embargo, es probable que desee las características que son exclusivas del plan Ai Ultra de Gemini, como la generación de videos VEO 3.

Ganador: Géminis


Plataformas

Puede acceder a ChatGPT y Gemini en la web o a través de aplicaciones móviles (Android e iOS). ChatGPT también tiene aplicaciones de escritorio (macOS y Windows) y una extensión oficial para Google Chrome. Gemini no tiene aplicaciones de escritorio dedicadas o una extensión de Chrome, aunque se integra directamente con el navegador.

Interfaz web chatgpt

(Crédito: OpenAI/PCMAG)

Chatgpt está disponible en otros lugares, Como a través de Siri. Como se mencionó, puede acceder a Gemini en las aplicaciones de Google, como el calendario, Documento, ConducirGmail, Mapas, Mantener, FotosSábanas, y Música de YouTube. Tanto los modelos de Chatgpt como Gemini también aparecen en sitios como la perplejidad. Sin embargo, obtiene la mayor cantidad de funciones de estos chatbots en sus aplicaciones y portales web dedicados.

Las interfaces de ambos chatbots son en gran medida consistentes en todas las plataformas. Son fáciles de usar y no lo abruman con opciones y alternar. ChatGPT tiene algunas configuraciones más para jugar, como la capacidad de ajustar su personalidad, mientras que la profunda interfaz de investigación de Gemini hace un mejor uso de los bienes inmuebles de pantalla.

Ganador: empate


Modelos de IA

ChatGPT tiene dos series primarias de modelos, la serie 4 (su línea de conversación, insignia) y la Serie O (su compleja línea de razonamiento). Gemini ofrece de manera similar una serie Flash de uso general y una serie Pro para tareas más complicadas.

Los últimos modelos de Chatgpt son O3 y O4-Mini, y los últimos de Gemini son 2.5 Flash y 2.5 Pro. Fuera de la codificación o la resolución de una ecuación, pasará la mayor parte de su tiempo usando los modelos de la serie 4-Series y Flash. A continuación, puede ver cómo funcionan estos modelos en una variedad de tareas. Qué modelo es mejor depende realmente de lo que quieras hacer.

Ganador: empate


ChatGPT y Gemini pueden buscar información actualizada en la web con facilidad. Sin embargo, ChatGPT presenta mosaicos de artículos en la parte inferior de sus respuestas para una lectura adicional, tiene un excelente abastecimiento que facilita la vinculación de reclamos con evidencia, incluye imágenes en las respuestas cuando es relevante y, a menudo, proporciona más detalles en respuesta. Gemini no muestra nombres de fuente y títulos de artículos completos, e incluye mosaicos e imágenes de artículos solo cuando usa el modo AI de Google. El abastecimiento en este modo es aún menos robusto; Google relega las fuentes a los caretes que se pueden hacer clic que no resaltan las partes relevantes de su respuesta.

Como parte de sus experiencias de búsqueda en la web, ChatGPT y Gemini pueden ayudarlo a comprar. Si solicita consejos de compra, ambos presentan mosaicos haciendo clic en enlaces a los minoristas. Sin embargo, Gemini generalmente sugiere mejores productos y tiene una característica única en la que puede cargar una imagen tuya para probar digitalmente la ropa antes de comprar.

Ganador: chatgpt


Investigación profunda

ChatGPT y Gemini pueden generar informes que tienen docenas de páginas e incluyen más de 50 fuentes sobre cualquier tema. La mayor diferencia entre los dos se reduce al abastecimiento. Gemini a menudo cita más fuentes que CHATGPT, pero maneja el abastecimiento en informes de investigación profunda de la misma manera que lo hace en la búsqueda en modo AI, lo que significa caretas que se puede hacer clic sin destacados en el texto. Debido a que es más difícil conectar las afirmaciones en los informes de Géminis a fuentes reales, es más difícil creerles. El abastecimiento claro de ChatGPT con destacados en el texto es más fácil de confiar. Sin embargo, Gemini tiene algunas características de calidad de vida en ChatGPT, como la capacidad de exportar informes formateados correctamente a Google Docs con un solo clic. Su tono también es diferente. Los informes de ChatGPT se leen como publicaciones de foro elaboradas, mientras que los informes de Gemini se leen como documentos académicos.

Ganador: chatgpt


Generación de imágenes

La generación de imágenes de ChatGPT impresiona independientemente de lo que solicite, incluso las indicaciones complejas para paneles o diagramas cómicos. No es perfecto, pero los errores y la distorsión son mínimos. Gemini genera imágenes visualmente atractivas más rápido que ChatGPT, pero rutinariamente incluyen errores y distorsión notables. Con indicaciones complicadas, especialmente diagramas, Gemini produjo resultados sin sentido en las pruebas.

Arriba, puede ver cómo ChatGPT (primera diapositiva) y Géminis (segunda diapositiva) les fue con el siguiente mensaje: “Genere una imagen de un estudio de moda con una decoración simple y rústica que contrasta con el espacio más agradable. Incluya un sofá marrón y paredes de ladrillo”. La imagen de ChatGPT limita los problemas al detalle fino en las hojas de sus plantas y texto en su libro, mientras que la imagen de Gemini muestra problemas más notables en su tubo de cordón y lámpara.

Ganador: chatgpt

¡Obtenga nuestras mejores historias!


Ícono del boletín


Toda la última tecnología, probada por nuestros expertos

Regístrese en el boletín de informes de laboratorio para recibir las últimas revisiones de productos de PCMAG, comprar asesoramiento e ideas.

Al hacer clic en Registrarme, confirma que tiene más de 16 años y acepta nuestros Términos de uso y Política de privacidad.

¡Gracias por registrarse!

Su suscripción ha sido confirmada. ¡Esté atento a su bandeja de entrada!


Generación de videos

La generación de videos de Gemini es la mejor de su clase, especialmente porque ChatGPT no puede igualar su capacidad para producir audio acompañante. Actualmente, Google bloquea el último modelo de generación de videos de Gemini, VEO 3, detrás del costoso plan AI Ultra, pero obtienes más videos realistas que con ChatGPT. Gemini también tiene otras características que ChatGPT no, como la herramienta Flow Filmmaker, que le permite extender los clips generados y el animador AI Whisk, que le permite animar imágenes fijas. Sin embargo, tenga en cuenta que incluso con VEO 3, aún necesita generar videos varias veces para obtener un gran resultado.

En el ejemplo anterior, solicité a ChatGPT y Gemini a mostrarme un solucionador de cubos de Rubik Rubik que resuelva un cubo. La persona en el video de Géminis se ve muy bien, y el audio acompañante es competente. Al final, hay una buena atención al detalle con el marco que se desplaza, simulando la detención de una grabación de selfies. Mientras tanto, Chatgpt luchó con su cubo, distorsionándolo en gran medida.

Ganador: Géminis


Procesamiento de archivos

Comprender los archivos es una fortaleza de ChatGPT y Gemini. Ya sea que desee que respondan preguntas sobre un manual, editen un currículum o le informen algo sobre una imagen, ninguno decepciona. Sin embargo, ChatGPT tiene la ventaja sobre Gemini, ya que ofrece un reconocimiento de imagen ligeramente mejor y respuestas más detalladas cuando pregunta sobre los archivos cargados. Ambos chatbots todavía a veces inventan citas de documentos proporcionados o malinterpretan las imágenes, así que asegúrese de verificar sus resultados.

Ganador: chatgpt


Escritura creativa

Chatgpt y Gemini pueden generar poemas, obras, historias y más competentes. CHATGPT, sin embargo, se destaca entre los dos debido a cuán únicas son sus respuestas y qué tan bien responde a las indicaciones. Las respuestas de Gemini pueden sentirse repetitivas si no calibra cuidadosamente sus solicitudes, y no siempre sigue todas las instrucciones a la carta.

En el ejemplo anterior, solicité ChatGPT (primera diapositiva) y Gemini (segunda diapositiva) con lo siguiente: “Sin hacer referencia a nada en su memoria o respuestas anteriores, quiero que me escriba un poema de verso gratuito. Preste atención especial a la capitalización, enjambment, ruptura de línea y puntuación. Dado que es un verso libre, no quiero un medidor familiar o un esquema de retiro de la rima, pero quiero que tenga un estilo de coohes. ChatGPT logró entregar lo que pedí en el aviso, y eso era distinto de las generaciones anteriores. Gemini tuvo problemas para generar un poema que incorporó cualquier cosa más allá de las comas y los períodos, y su poema anterior se lee de manera muy similar a un poema que generó antes.

Recomendado por nuestros editores

Ganador: chatgpt


Razonamiento complejo

Los modelos de razonamiento complejos de Chatgpt y Gemini pueden manejar preguntas de informática, matemáticas y física con facilidad, así como mostrar de manera competente su trabajo. En las pruebas, ChatGPT dio respuestas correctas un poco más a menudo que Gemini, pero su rendimiento es bastante similar. Ambos chatbots pueden y le darán respuestas incorrectas, por lo que verificar su trabajo aún es vital si está haciendo algo importante o tratando de aprender un concepto.

Ganador: chatgpt


Integración

ChatGPT no tiene integraciones significativas, mientras que las integraciones de Gemini son una característica definitoria. Ya sea que desee obtener ayuda para editar un ensayo en Google Docs, comparta una pestaña Chrome para hacer una pregunta, pruebe una nueva lista de reproducción de música de YouTube personalizada para su gusto o desbloquee ideas personales en Gmail, Gemini puede hacer todo y mucho más. Es difícil subestimar cuán integrales y poderosas son realmente las integraciones de Géminis.

Ganador: Géminis


Asistentes de IA

ChatGPT tiene GPT personalizados, y Gemini tiene gemas. Ambos son asistentes de IA personalizables. Tampoco es una gran actualización sobre hablar directamente con los chatbots, pero los GPT personalizados de terceros agregan una nueva funcionalidad, como el fácil acceso a Canva para editar imágenes generadas. Mientras tanto, terceros no pueden crear gemas, y no puedes compartirlas. Puede permitir que los GPT personalizados accedan a la información externa o tomen acciones externas, pero las GEM no tienen una funcionalidad similar.

Ganador: chatgpt


Contexto Windows y límites de uso

La ventana de contexto de ChatGPT sube a 128,000 tokens en sus planes de nivel superior, y todos los planes tienen límites de uso dinámicos basados ​​en la carga del servidor. Géminis, por otro lado, tiene una ventana de contexto de 1,000,000 token. Google no está demasiado claro en los límites de uso exactos para Gemini, pero también son dinámicos dependiendo de la carga del servidor. Anecdóticamente, no pude alcanzar los límites de uso usando los planes pagados de Chatgpt o Gemini, pero es mucho más fácil hacerlo con los planes gratuitos.

Ganador: Géminis


Privacidad

La privacidad en Chatgpt y Gemini es una bolsa mixta. Ambos recopilan cantidades significativas de datos, incluidos todos sus chats, y usan esos datos para capacitar a sus modelos de IA de forma predeterminada. Sin embargo, ambos le dan la opción de apagar el entrenamiento. Google al menos no recopila y usa datos de Gemini para fines de capacitación en aplicaciones de espacio de trabajo, como Gmail, de forma predeterminada. ChatGPT y Gemini también prometen no vender sus datos o usarlos para la orientación de anuncios, pero Google y OpenAI tienen historias sórdidas cuando se trata de hacks, filtraciones y diversos fechorías digitales, por lo que recomiendo no compartir nada demasiado sensible.

Ganador: empate

Continue Reading

Trending

Noticias de Inteligencia Artificial
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.