OpenAI’s Deep Research can save you hours of work – and now it’s a lot cheaper to access

What’s higher than an AI chatbot that may help you with duties? One that may do them for you. OpenAI continues to construct out its AI brokers in ChatGPT with the launch of Deep Analysis. 

Deep Analysis

Earlier this month, OpenAI unveiled Deep Analysis, an AI agent that may conduct multi-step analysis for you by pulling a sturdy quantity of knowledge from the online and synthesizing these sources for you in a complete report. As soon as prompted, Deep Analysis can work totally independently; it is like having a analysis analyst at your command. 

At launch, it was solely accessible to ChatGPT Professional customers, which means you would need to pay $200 monthly. Now, Deep Analysis is rolling out to extra paid subscribers, together with ChatGPT Plus, Crew, Edu, and Enterprise customers. Nevertheless, the added customers can have 10 deep analysis queries monthly, in comparison with the 120 deep analysis queries monthly Professional customers have entry to.

Powering Deep Analysis is a model of the OpenAI o3 mannequin optimized for net looking and information evaluation. By leveraging o3’s superior reasoning capabilities, it may possibly search and interpret large quantities of content material from the online, together with texts, photos, and extra, after which output it in a report focused to your wants. 

Every report is generated in 5 to half-hour, relying on the duty at hand. Nevertheless, you’ll be able to work on different duties throughout that point, optimizing your workflow productiveness. The completed report is output within the chat. Together with the replace that expanded entry to the characteristic, OpenAI introduced deep analysis now consists of embedded photos with citations within the output and is healthier at understanding uploaded information. 

In line with OpenAI, the identical work would take people hours. Moreover, the agent is supposed to be significantly good at discovering area of interest data that may require people to carry out a number of searches.

The audience for Deep Analysis consists of those that do intensive data work in finance, science, coverage, and engineering — and who want dependable, thorough analysis, in keeping with OpenAI. Each report consists of clear citations and a abstract of the agent’s pondering in order that customers can double-check the knowledge for themselves. 

Double-checking a chatbot’s responses is usually good apply, as chatbots are liable to hallucinations. Particularly, OpenAI warns that Deep Analysis “can generally hallucinate info in responses or make incorrect inferences, although at a notably decrease fee than current ChatGPT fashions, in keeping with inside evaluations.” OpenAI additionally added that the agent might wrestle to tell apart authoritative data from rumors and might fail to convey uncertainty accurately, highlighting the necessity for human evaluate. 

Efficiency in contrast

Within the weblog submit saying the characteristic, OpenAI consists of the identical side-by-side outcomes of GPT-4o versus Deep Analysis to showcase how the identical immediate generates very completely different outcomes. Those generated with Deep Analysis had been rather more sturdy and higher organized. 

Deep Analysis additionally outperformed GPT-4o on Humanity’s Final Examination, a just lately launched AI benchmark examination by Scale AI and the Middle for AI Security (CAIS) that checks numerous topics on expert-level questions. Deep Analysis scored a 26.6% accuracy, outperforming GPT-4o, Grok-2, Claude 3,5 Sonnet, Gemini Pondering, o1, and even o3-mini excessive, which had simply scored the very best rating a few days prior, as highlighted by OpenAI CEO Sam Altman.  

OpenAI additionally printed Deep Analysis’s efficiency outcomes on a collection of different evaluations, together with GAIA⁠, a public benchmark that evaluates AI on real-world questions and an inside analysis of expert-level duties throughout completely different areas of deep analysis. In each, Deep Analysis had spectacular outcomes, even topping the GAIA exterior leaderboard. 

Alternate options

If you need entry to the characteristic now however do not wish to shell out the $200 monthly, Google has an analogous characteristic, additionally referred to as Deep Analysis, that’s out there to all of its Gemini Superior customers by the Google One AI Premium plan that prices $20 monthly. 

Again in December, Altman even replied to an X person who requested Altman to “do a deep analysis characteristic like Gemini however higher,” with “kk,” suggesting that the newly launched Deep Analysis characteristic is OpenAI’s reply to Google. 

Not too long ago, xAI additionally launched its personal AI analysis agent, referred to as DeepSearch. X Premium and Premium+ customers have entry to DeepSearch now, with all capabilities being rolled our to all Grok customers with limits. The X Premium membership is $8 monthly or $84 per yr, whereas the Premium+ membership is $40 monthly or $395 per yr, in keeping with X. 

Microsoft additionally introduced a characteristic able to extra thorough reasoning referred to as Assume Deeper, which permits customers to leverage OpenAI’s O1 reasoning mannequin to ship higher-quality responses to complicated prompts. Nevertheless, not like Gemini, OpenAI’s and Grok’s brokers, it does not have agentic capabilities or entry to the web. The largest perk is that the expertise is totally free.