Latest / How I Work / How I condensed a day of research into 5 minutes (without losing quality).
Transcript
- 0:00Speaker 1: If you've ever fallen down a Google rabbit hole trying
- 0:04Speaker 1: to prepare for a workshop, or write a strategy paper,
- 0:08Speaker 1: or just get your head around a new topic, you
- 0:10Speaker 1: are going to love this episode. Today, Neo Applan, Inventium's
- 0:16Speaker 1: Jenai Guru, is back to explain one of Jenai's most
- 0:21Speaker 1: powerful hidden tools, deep research. It's like outsourcing four hours
- 0:27Speaker 1: of online digging in five minutes. Welcome to How I Work,
- 0:36Speaker 1: a show about habits, rituals, and strategies for optimizing your day.
- 0:41Speaker 1: I'm your host, Doctor Amantha Imber. Before we get into
- 0:45Speaker 1: today's episode, I have been geeking out over the research
- 0:48Speaker 1: on AI adoption in the workplace, and the data is
- 0:51Speaker 1: really clear. The people who master Jenai now aren't just
- 0:56Speaker 1: getting ahead, they are creating entirely new level of productivity
- 1:01Speaker 1: that seemed impossible just like a year ago. Because in
- 1:05Speaker 1: a nutshell, there are two types of professionals, those who
- 1:09Speaker 1: are fluent in AI and those left scrambling to catch up.
- 1:14Speaker 1: Inventium's Genai Productivity Upgrade is how you can make sure
- 1:18Speaker 1: that you are in that first group. It is a
- 1:21Speaker 1: twelve week course designed to move you from dabbler to
- 1:23Speaker 1: productivity machine. No fluff, just practical strategies that will pay
- 1:27Speaker 1: off from week one, saving you at least ten hours
- 1:31Speaker 1: every single week. You'll learn how to make AI sound
- 1:34Speaker 1: exactly like you use AI as your second brain to
- 1:37Speaker 1: excel at your job and use it to become a
- 1:40Speaker 1: much more effective leader, and so much more. Whether you're
- 1:45Speaker 1: a complete beginner or already dabbling, we've got you covered,
- 1:48Speaker 1: starting with prompting fundamentals and going all the way through
- 1:52Speaker 1: to advanced automations and digentic AI week kickoff on July fourteen,
- 1:57Speaker 1: and spots are limited. Visit inventi do au forward slash
- 2:02Speaker 1: Jenai hyphen cohort to secure your place now and there's
- 2:05Speaker 1: a link to that in the show notes. You've got
- 2:08Speaker 1: nothing to lose because there's a seven day money back guarantee,
- 2:12Speaker 1: so head to the show notes click on the link
- 2:14Speaker 1: to read more about the program today. Okay, neo, what
- 2:19Speaker 1: exactly is deep research and what problem is it solving? Oh?
- 2:24Speaker 2: You know how when you've got to do a lot
- 2:26Speaker 2: of research and you think I'm going to have to
- 2:27Speaker 2: spend a couple of hours working with Google to learn
- 2:30Speaker 2: about this thing, to research about this thing, so it
- 2:33Speaker 2: might be a lot of manual research that you need
- 2:36Speaker 2: to do instead of doing that manually, which can take
- 2:39Speaker 2: many hours, so you can learn the topic and know
- 2:41Speaker 2: that you're getting the right information. You can automate that
- 2:45Speaker 2: with using Jenai tools and the deep research function. So
- 2:49Speaker 2: it's great if you want to do things like market trends,
- 2:52Speaker 2: best practices, science updates and all those kind of things. Effectively,
- 2:56Speaker 2: you're getting the ai to be your research assistant and
- 2:59Speaker 2: it goes away. It plans what the research is going
- 3:01Speaker 2: to be, and it goes and searches so many websites
- 3:04Speaker 2: for you.
- 3:05Speaker 1: Okay, So which genai tools actually offer deep research and
- 3:11Speaker 1: how do they differ?
- 3:13Speaker 2: The main functions are pretty much the same between the
- 3:16Speaker 2: different ais. So chatchpt does it on the paid and
- 3:20Speaker 2: the free version. You get five deep researchers a month
- 3:23Speaker 2: on the free and thirty a month on the paid.
- 3:27Speaker 2: Perplexity has it, as well as Gemini, which is Google's
- 3:31Speaker 2: Genai tool. And yeah, they're all very good. They're slightly
- 3:35Speaker 2: different in the way that they approach the results. However,
- 3:38Speaker 2: I'd say play with one many of them. They're awesome
- 3:41Speaker 2: at what they do.
- 3:43Speaker 1: Do you have a personal.
- 3:44Speaker 2: Favorite At the moment, I'm enjoying Gemini, and the reason
- 3:48Speaker 2: for that is that I can use it as much
- 3:51Speaker 2: as I want to. It's free at the moment. So
- 3:54Speaker 2: I do many deep researchers, but look, all of them
- 3:58Speaker 2: are excellently in the way that they operate.
- 4:00Speaker 1: And I think what's useful with Gemini If you are
- 4:03Speaker 1: on the Google ecosystem, it exports the deep research report
- 4:07Speaker 1: to Google Docs and then chat, GPT and Perplexity export
- 4:10Speaker 1: to word and PDF, so that is useful as well.
- 4:13Speaker 1: Now I want to know what is actually happening in
- 4:17Speaker 1: the background when AI is doing its deep research.
- 4:21Speaker 2: It's a true agentic flow and I'm doing like air
- 4:24Speaker 2: quotes here agentic, so gentic is where the AI is
- 4:27Speaker 2: going off and doing things on its own. So it's
- 4:30Speaker 2: great where the interface actually shows you what the AI
- 4:33Speaker 2: is thinking about in order to do that deep research.
- 4:36Speaker 2: So it will say these are the sources that they
- 4:38Speaker 2: need to look at. Then I need to look at this,
- 4:40Speaker 2: Then I need to research this particular thing, and it
- 4:43Speaker 2: will then plan how it's going to approach this problem
- 4:46Speaker 2: and all the different types of pages it's going to search.
- 4:49Speaker 2: So behind the scenes, it's doing the plan first. Now
- 4:52Speaker 2: before you get it to do the work. You can
- 4:54Speaker 2: agree with that plan or not, and you could say
- 4:56Speaker 2: I need to do more of this than that. So
- 4:58Speaker 2: work with it first the plan, and then when you
- 5:01Speaker 2: hear the go button on the deep research button, then
- 5:04Speaker 2: it does exactly what it planned to. So it'll go
- 5:06Speaker 2: and research many different websites. So I've had ones where
- 5:09Speaker 2: it's like one hundred and fifty websites kind of common.
- 5:12Speaker 2: I've had some up to four hundred and ninety different
- 5:15Speaker 2: websites that it's looking at in order for it to
- 5:17Speaker 2: go through the plan. So it gives you so much
- 5:21Speaker 2: information from those sites. And then what it does is
- 5:24Speaker 2: it builds you a report. And the report is very detailed.
- 5:28Speaker 2: We're talking on it five ten pages worth of information.
- 5:31Speaker 2: Plus it also sites where it got the information for
- 5:34Speaker 2: so you can click on the links and learn more
- 5:37Speaker 2: and just make sure that it's got it right. So
- 5:39Speaker 2: in summary, it plans the search, work with it on
- 5:41Speaker 2: the plan. It then does the search, and it produces
- 5:45Speaker 2: you an excellent report, so you've got an outcome.
- 5:47Speaker 1: Okay, So something I have wondered about is how exactly
- 5:52Speaker 1: should I best write the prompt to get the best
- 5:55Speaker 1: result from deep research? Because I know that in the
- 5:58Speaker 1: past I've sometimes been quite lazy. I've just written a
- 6:01Speaker 1: quick one sentence prompt. I mean, the results have been
- 6:05Speaker 1: pretty good, But ideally, what should I be doing with
- 6:08Speaker 1: the prompt?
- 6:09Speaker 2: Same as any other prompt, If it knows what you're
- 6:12Speaker 2: trying to achieve, and it knows what the goals are
- 6:14Speaker 2: and the context and things like that, you will get
- 6:16Speaker 2: a much better result. So instead of saying what's the
- 6:20Speaker 2: modern strategy for banking or something like that, very generic,
- 6:25Speaker 2: you need to say what specific strategy you're looking at.
- 6:28Speaker 2: So this is the actual area I'm looking at, and
- 6:30Speaker 2: these are the types of things I'm trying to solve.
- 6:32Speaker 2: So work with the AI on this. You can even
- 6:36Speaker 2: do what they call meta prompting, which is working with
- 6:38Speaker 2: the AI giving it all the context and what you're
- 6:41Speaker 2: trying to achieve and then get it to produce you
- 6:43Speaker 2: a prompt, which is a nice little cheats way. But
- 6:46Speaker 2: the prompt is not the most important part. It is
- 6:50Speaker 2: equally as important as working with AI on that search strategy,
- 6:54Speaker 2: So making sure it's going to search the right kind
- 6:56Speaker 2: of pages and trying to get the right kind of
- 6:58Speaker 2: outcomes in the two of them together equally is important
- 7:01Speaker 2: for a great result.
- 7:03Speaker 1: And what's your advice on working with the AI to
- 7:05Speaker 1: design a good approach to doing the deep research?
- 7:10Speaker 2: Saying my goals are and here's the context, like it's
- 7:14Speaker 2: the business, here's what I'm trying to achieve. These are
- 7:16Speaker 2: the two main things that it really needs to know
- 7:19Speaker 2: to be able to nail a great deep research prompt
- 7:22Speaker 2: or a deep research process. So once you've got those
- 7:25Speaker 2: things absolutely agreed with the AI, then you'll find you're
- 7:28Speaker 2: going to get a much better result. So better background
- 7:31Speaker 2: and better context you can give the better the results.
- 7:34Speaker 1: Okay, Now, a question that I'm sure a lot of
- 7:37Speaker 1: listeners are asking in their brains is can we trust
- 7:42Speaker 1: the results? I mean, we all know about hallucinations, where
- 7:46Speaker 1: the AI just makes stuff up, it creates sources out
- 7:50Speaker 1: of nothing. Is deep research different in that sense?
- 7:54Speaker 2: It isn't in that you can get hallucinations. However, I
- 7:58Speaker 2: found that there are few hallucinations, and the reason for
- 8:01Speaker 2: that is it's effectively summarizing web pages that it has
- 8:05Speaker 2: already just read. It's not hallucinating out of its own knowledge.
- 8:08Speaker 2: And if there's a knowledge gap, there's a problem. And
- 8:11Speaker 2: so out of the box you will find that there
- 8:13Speaker 2: are fewer hallucinations. The worst that I've found is it's
- 8:17Speaker 2: interpreted a page slightly differently than I would have interpreted it.
- 8:22Speaker 2: And the page may have been for a tangentially related
- 8:25Speaker 2: problem or industry, and I can see how it's got
- 8:28Speaker 2: the things kind of together, but in reality, I would
- 8:31Speaker 2: have said, yeah, probably that page wouldn't have been something
- 8:33Speaker 2: that I would have included on my search. So it's
- 8:37Speaker 2: very accurate. It's more of an interpretation issue that I
- 8:40Speaker 2: have found, and even then it's really rare for it
- 8:43Speaker 2: of me to have found it after all the deep
- 8:45Speaker 2: researches that I've done, so it is a very thorough search.
- 8:50Speaker 2: It is really good at summarizing the pages that it
- 8:52Speaker 2: has read, and even if it has read fo one
- 8:55Speaker 2: hundred and eighty pages do notes, so just because it's
- 8:57Speaker 2: done a deep research that it isn't the world's expert
- 9:00Speaker 2: on those things. So it's just summarized the four hundred
- 9:04Speaker 2: odd pages that it's read, and it's done an awesome
- 9:06Speaker 2: job at the summarization. However, you might need to still
- 9:10Speaker 2: do some more research on top of it so know
- 9:12Speaker 2: that it could potentially hallucinate. It also might not have
- 9:16Speaker 2: searched a really important page that's out there, just as
- 9:19Speaker 2: you and I when we do a Google search, we
- 9:21Speaker 2: may not have searched every paper out there. So the
- 9:24Speaker 2: probably the biggest risk, biggest takeaway is review everything. Click
- 9:28Speaker 2: on it, and if you're wanting to know more about it,
- 9:30Speaker 2: click on those links so you can be sure that
- 9:32Speaker 2: everything in there is correct.
- 9:34Speaker 1: Neo, thank you so much for taking us through deep research.
- 9:39Speaker 1: It is hopefully a function in people's GENAI that they
- 9:43Speaker 1: are going to now use a lot more frequently. If
- 9:48Speaker 1: you like today's jow, make sure you follow on your
- 9:51Speaker 1: podcast app to be alerted when new episodes drop. How
- 9:55Speaker 1: I Work was recorded on the traditional land of the
- 9:58Speaker 1: Warrangery People, part of the Cool and Nake. A big
- 10:01Speaker 1: thank you to Martin Imma for doing the sound mix.